21st ANNUAL AUSTRALIAN POULTRY SCIENCE SYMPOSIUM
SYDNEY, NEW SOUTH WALES
1 -3RD FEBRUARY 2010
Organised by
THE POULTRY RESEARCH FOUNDATION
(University of Sydney)
and
THE WORLD’S POULTRY SCIENCE ASSOCIATION
(Australian Branch)
Papers presented at this Symposium have been refereed by external referees and by members of the Editorial
Committee. However, the comments and views expressed in the papers are entirely the responsibility of the
author or authors concerned and do not necessarily represent the views of the Poultry Research Foundation or the
World’s Poultry Science Association.
Enquiries regarding the Proceedings should be addressed to:
The Director, Poultry Research Foundation
Faculty of Veterinary Science, University of Sydney
Camden NSW 2570
Tel:
Fax:
02 46 550 656; 9351 1656
02 46 550 693; 9351 1693
ISSN-1034-6260
AUSTRALIAN POULTRY SCIENCE SYMPOSIUM
2010
ORGANISING COMMITTEE
Dr. Peter Groves (Acting Director)
Dr P. Selle (Editor)
Ms. L. Browning (President PRF)
Ms. M. Betts
Professor W.L. Bryden
Dr. D. Cadogan
Dr. A. Crossan
Mr. P. Doyle
Professor D.J. Farrell
Mr. G. Hargreave
Dr. R. Hughes
Dr. A. Kocher
Mr. G. McDonald
Dr. W. Muir
Ms. J. O’Keeffe
Dr. R. Pym
Dr. J. Roberts
Dr T.M. Walker
The Committee thanks the following, who refereed papers for the Proceedings:
D.J. Cadogan
G.M. Cronin
J.A. Downing
P.J. Groves
R.J. Hughes
W.I. Muir
R. Ravindran
P.H. Selle
T. Walker
The Committee would also like to recognise the following Chairpersons for their contribution
to:
Australian Poultry Science Symposium 2009
Ms. Linda Browning – President - Poultry Research Foundation
Professor Julie Roberts – President - Australian WPSA Branch
Dr. Peter Selle – University of Sydney
Dr. Jeff Downing – University of Sydney
Professor Mingan Choct – Australian Poultry CRC
Dr. Vivien Kite – RIRDC Chicken Meat Programme
Dr. Ron MacAlpine – Inghams Enterprises
Dr. Angus Crossan – AECL
Mr. Greg Hargreave – Baiada Poultry Pty. Ltd
Dr. Peter Groves – Director - Poultry Research Foundation
AUSTRALIAN POULTRY AWARD
The Australian Poultry Award is presented annually to an Australian resident who has made a
long-term outstanding contribution to poultry science and/or the Australian poultry industry.
The Award is made by the Australian Branch of the World’s Poultry Science Association
(WPSA) and takes the form of a suitably inscribed plaque which includes the winner’s name,
together with a framed citation. Nominations are called for early each year from the
membership of WPSA, and completed nominations require to be forwarded to the Secretary
of the Australian Branch no later than 31st July. The selection committee consists of the
Australian Branch Management Committee of WPSA (10 members) as well as Award
recipients from the previous 10 years who are still active in the Australian poultry Industry.
Voting is by secret postal ballot, and if more than two candidates are nominated, a preferential
voting system is used. The Award is made to the winner at suitable forums where poultry
industry people are gathered, such as the annual Australian Poultry Science Symposium, the
biennial Poultry Information Exchange (PIX), and the triennial Australian Poultry
Convention.
Previous recipients of the award are:
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
Mr A.O. Moll
Dr M.W. McDonald
Professor R.B. Cumming
Mr F. Skaller
Professor G.L. McClymont
Dr S. Hunt
Dr L. Hart
Mr N. Milne
Mr R. Morris
Mr J. & Mr R. Ingham
Mr S.J. Wilkins
Professor C.G. Payne
Mr W. Stanhope
Professor B. Sinkovic
Mr J. Douglas
Mr D. Blackett
Dr A.F. Webster
Mr R. Fuge
Dr J.G. Fairbrother
Dr R.K. Ryan
Mr C. Donnelley
Dr P. Gilchrist
Dr C.A.W. Jackson
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
Mr E. Rigby
Mr W. Shaw
Dr H. Bray
Dr M. Mackenzie
Professor D.J. Farrell
Dr B.L. Sheldon
Mr R. Macindoe
Mr B. Bartlett
Dr R.A.E. Pym
Dr E.E. Best
Mr M. Peacock
Professor D. Balnave
Dr H. Westbury
Mr L. Brajkovich
Mr R.J. Hughes
Dr T.M. Grimes
Dr R. MacAlpine
Dr M. Choct
Professor P. Spradbrow
Dr Juliet R.Roberts
Dr Vivien Kite
Mr Rowly Horn
SPONSORS of the 2010
AUSTRALIAN POULTRY SCIENCE SYMPOSIUM
Invited Speaker Sponsors
AECL Egg Program
Poultry Research Foundation
RIRDC Chicken Meat Program
Gold Sponsors
Australian Poultry CRC
DSM Nutritional Products Pty. Ltd
Feedworks / Danisco
Phibro Animal Health / Alpharma Animal Health
Silver Sponsors
ADM Australia Pty. Ltd
Alltech Biotechnology Pty. Ltd
Evonik Degussa Australia Pty. Ltd
Bronze Sponsors
Baiada Poultry Pty. Ltd
BEC Feed Solutions
Biomin Australia
Elanco Animal Health
International Animal Health
JEFO Australia
Kemin (Australia)
Alternative Sponsors
Kemira
Zootechny Pty. Ltd
CONTENTS
RECOGNITION OF THE 100TH ANNIVERSARY OF THE FACULTY OF VETERINARY
SCIENCE – UNIVERSITY OF SYDNEY
i
JOHN.L.BARNETT IN MEMORIAM
v
INNOVATIONS IN EUROPEAN POULTRY PRODUCTION
DEVELOPMENTS AND INNOVATIONS IN BROILER NUTRITION IN THE NETHERLANDS
L.L. de Lange – De Heus Voeders B.V. The Netherlands
1
CHALLENGING CURRENT POULTRY FEEDING DOGMAS BY FEED INTAKE RESTRICTION
9
AND THE USE OF COARSE FEED INGREDIENTS
B. Svihus - Norwegian University of Life Sciences Norway
CHALLENGES TO EUROPEAN POULTRY PRODUCTION
JD. van der Klis - Schothorst Feed Research The Netherlands
17
UPDATE ON CURRENT EUROPEAN BROILER BONE PROBLEMS
C. Whitehead – Roslin Institute UK
22
NUTRITION SYSTEMS BEYOND ME
UTILISATION OF METABOLISABLE ENERGY OF FEEDS IN PIGS AND POULTRY: INTEREST
OF NET ENERGY SYSTEMS?
J. Noblet, J. van Milgen and S. Dubois – INRA France
26
AN EFFECTIVE ALTERNATIVE TO THE METABOLISABLE ENERGY SYSTEM
R. Gous – University of KwaZulu-Natal South Africa
36
ENERGY IN POULTRY DIETS: ADJUSTED AME OR NET ENERGY
J D. van der Klis, C. Kwakernaak, A. Jansman and M. Blok - Schothorst Feed
Research The Netherlands
44
NON-STARCH POLYSACCHARIDES AND ENZYME APPLICATION INFLUENCE THE NET
50
ENERGY VALUE OF BROILER DIETS
M. Choct, A. Tukei and D.J. Cadogan – Australian Poultry CRC
BROILER NUTRITION
UPDATE ON NEAR INFRARED REFLECTANCE ANALYSIS OF GRAINS TO ESTIMATE
51
NUTRITIONAL VALUE FOR CHICKENS
J.L. Black, R.J. Hughes, M.S. Geier, S.G. Nielsen, A.M. Tredrea and P.C. Flinn – J.
Black Consulting
THE PERFORMANCE OF BROILERS OFFERED SORGHUM-COMPARED TO WHEAT-BASED
DIETS: A LARGE SCALE EXPERIMENT
R.A. Perez-Maldonado and H. Rodrigues – Dept. Primary Industries & Fisheries
QLD
55
BROILER NUTRITION (Cont.)
ENERGY ULITISATION AND DIGESTIBILITY OF FATS AS INFLUENCED BY THE AGE OF
59
BROILERS
P. Tancharoenrat, F. Zaefarian, G. Ravindran and R. Ravindran, – Massey
University NZ
TOOLS IN EARLY NUTRITION TO MAXIMISE GROWTH PERFORMANCE
A.Kocher, A. Naylor, C. Martin, T. Wilson and J. Hazeldene - Alltech
Biotechnology
60
DIETS HIGH IN LINOLEIC ACID REDUCE OMEGA-3 LONG CHAIN POLYUNSATURATED
64
FATTY ACIDS IN CHICKEN TISSUE
L.R. Kartikasari, R. J. Hughes, M.S. Geier, M. Makrides and R.A. Gibson –
University of Adelaide
THE VULNERABILITY OF SORGHUM TO “MOIST-HEAT”
P.H. Selle, R.J. Gill and J.A. Downing – University of Sydney
68
THE NUTRITIVE VALUE OF HIGH-YIELDING TRITICALE VARIETIES AND THEIR
72
POTENTIAL FOR INCLUSION IN POULTRY DIETS
A.Widodo, P.Iji and J.V. Nolan – University of New England
EFFECTS OF HOUSING AND LITTER ON BEHAVIOUR AND
FOOD QUALITY
INFLUENCE OF HOUSING SYSTEMS ON THE BACTERIOLOGOICAL QUALITY AND SAFETY
OF TABLE EGGS
74
K. de Reu, W. Messens, K. Gruspeerdt, M. Heyndrickx, B. Rodenburg, M.
Uyttendaele and L.Herman – Institute for Agriculture and Fisheries Research
(ILVO)
LITTER CONSUMPTION BY POULTRY AS AFFECTED BY DIET STRUCTURE
B. Svihus and H. Hetland - Norwegian University of Life Sciences Norway
82
THE EFFECTS OF INTERRUPTING A DUSTBATHING BOUT ON THE CHOICE BEHAVIOUR OF
LAYING HENS IN A Y-MAZE TEST
S.M. Laine, G.M. Cronin, J.C. Petherick and P.H. Hemsworth – University of
Melbourne
85
EGG QUALITY IN THE AUSTRALIAN EGG INDUSTRY: AN UPDATE
J.R. Roberts and K.K. Chousalkar – Australian Poultry CRC
86
BROILER PERFORMANCE AND NUTRITION
OPTIMAL SULPHUR AMINO ACIDS TO LYSINE RATION IN GROWER PHASE IN ROSS 308
90
BROILERS
T.G. Madsen, E. Hangoor, P.J.A. Wijtten, J.K.W.M. Sparrla and A. Lemme –
Evonik Degussa Pty. Ltd
DIETARY ENZYMES ALTER SORGHUM PROTEIN DIGESTIBILITY AND AME CONTENT
A.Sultan, X. Li, D. Zhang, D.J. Cadogan and W.L. Byrden - University of
Queensland
94
BROILER PERFORMANCE AND NUTRITION (Cont.)
INFLUENCE OF ENERGY AND BALANCED PROTEIN LEVELS IN WHEAT-BASED DIETS ON
ROSS 308 BROILER PERFORMANCE
T.G. Madsen, S. Carroll, C. Kemp and A. Lemme – Evonik Degussa Pty Ltd
VARIATION IN NUTRIENT COMPOSITION AND STRUCTURE OF HIGH-MOISTURE MAIZE
DRIED AT DIFFERENT TEMPERATURES
95
99
M.M. Bhuiyan, P.A. Iji, A.F. Islam and L.L. Mikkelsen – Unversity of New England
INFLUENCE OF FEED FORM ON INTAKE PREFERENCE AND PERFORMANCE OF YOUNG
BROILERS
103
K.H. Huang and M. De Beer – Aviagen Inc. USA
HEALTH
DIFFERENTIATION BETWEEN PATHOGENIC SEROTYPE 1 ISOLATES OF MAREK’S DISEASE
VIRUS (MDV1) AND THE RISPENS VACCINE IN AUSTRALIA USING REAL-TIME PCR.
K.G. Renz, B.F. Cheetham and S.W. Walkden-Brown - University of New England
107
A QUANTITATIVE PROFILE OF INFECTIOUS BRONCHITIS VIRUS IN FAECES OF LAYING
HENS
111
K.K. Chousalkar and J.R. Roberts – Charles Stuart University
THE STRATEGIC USE OF ORGANIC ACIDS TO IMPROVE GUT HEALTH IN POULTRY
L. Li - Kemira Asia Pacific, Singapore
115
INACTIVATION OF VIRUSES AND COCCIDIA IN BROILER LITTER FOLLOWING HEAPING
118
OR WINDROWING AT THE END OF THE BATCH
A.F.M.F. Islam, S.K. Burgess, P.Easey, B. Wells and S.W. Walkden-Brown –
University of New England
SPATIAL AND TEMPORAL VARIATION IN AMMONIA CONCENTRATIONS IN BROILER
SHEDS: EFFECTS OF CHICKEN AGE AND SHED TYPE
A.F.M.F. Islam, M. Dunlop, B. Wells and S.W. Walkden-Brown – University of New
England
122
LAYER NUTRITION, WELFARE AND FOOD SAFETY
EGGSHELL FACTORS INFLUENCING EGGSHELL PENETRATION AND WHOLE EGG
CONTAMINATION BY DIFFERENT BACTERIA, INCLUDING SALMONELLA ENTERITIDIS
K. de Reu, W. Messens, K. Gruspeerdt, M. Heyndrickx, M. Uyttendaele and L.
Herman - Institute for Agriculture and Fisheries Research (ILVO)
126
THE EFFECTS OF TWO LIGHT-DARK SCHEDULES ON EGG LAYING TIME AND
SYNCHRONY, THE INCIDENCE OF LAYING IN THE DARK AND HEN WELFARE
130
G.M. Cronin, S.S. Borg, T.H. Storey, J.A. Downing and J.L. Barnett – University of
Sydney
ATTRACTING LAYING HENS INTO RANGE AREAS USING SHELTERBELTS
E.A. Borland, S. Hazel, P.C. Glatz, B.K. Rodda, H. Rimmington, S.C. Wyatt and
Z.H. Miao – University of Adelaide
134
LAYER NUTRITION, WELFARE AND FOOD SAFETY (Cont.)
ATTRACTING LAYING HENS INTO RANGE AREAS USING SHADE AND FORAGE
P.C. Glatz, B.K. Rodda, H. Rimmington, S.C. Wyatt and Z.H. Miao - SARDI
135
OPTIMUM LEVEL OF CASSAVA PULP IN DIETS FOR LAYERS
N. Chauynarong, P.A. Iji and U. Kanto – University of New England
136
IMPLICATIONS ON THE USE OF CANOLA MEAL AND RECOMMENDATIONS TO REDUCE
FISHY TAINT IN EGGS FROM BROWN HENS
140
R. Perez-Maldonado and T. Treloar – Rider Perez-Maldonado Consultancy
EFFECT OF A COMPLEX ENZYME ON THE DIGESTIBILITY OF LAYER DIETS CONTAINING
PALM KERNEL MEAL
141
V. Ravindran, D. Thomas, A. Leary and A.Kocher - Alltech Biotechnology
GLOBAL GOALS TO PROMOTE POULTRY PRODUCTION IN
DEVELOPING COUNTRIES
OPPORTUNITIES AND SUSTAINABILITY OF SMALLHOLDER POULTRY PRODUCTION IN
145
THE SOUTH PACIFIC REGION
P. Glatz, I.D. Black, W. Ayalew, J.K. Pandi, R.J. Hughes, Z.H. Miao, J. Wahanui, T.
Jansen, V. Manu and B.K. Rodda – SARDI
THE ROLE OF THE WORLD’S POULTRY SCIENCE ASSOCIATION (WPSA) IN SUPPORT OF
POULTRY PRODUCTION IN DEVELOPING COUNTRIES
153
R. Pym - WPSA
THE EVOLVING ROLE OF FAO IN POULTRY DEVELOPMENT
S.D. Mack – FAO Italy
161
IMPROVED VILLAGE POULTRY PRODUCTION IN THE SOLOMON ISLANDS
R. Parker – Kai Kokorako Perma-Poultry
171
HEAVY METAL CONTAMINATION IN MINERAL SOURCES FOR MONOGASTRIC FEED IN
174
ASIA PACIFIC
T. Jarman, A. Frio, A.Leary, A.Kocher, S. Fike and B. Timmons - Alltech AsiaPacific Bioscience Centre
COMMERCIAL DUCK PRODUCTION
DUCK PRODUCTION IN AUSTRALIA
P.R. Brown - Pepe’s Ducks
178
THE EFFECTS OF STRAIN AND SEASON ON THE PERFORMANCE OF COMMERCIAL DUCKS
UNDER AUSTRALIAN CONDITIONS
J.A. Downing and W. Taylor - University of Sydney
182
THE PERFORMANCE OF COMMERCIAL DUCKS FED DIFFERENT DIETARY PROTEIN
186
CONCENTRATIONS DURING THE FINISHER PHASE
J.A. Downing and J. Ischan – University of Sydney
TWO-DAY-OLD DUCKLINGS INTERACT MORE WITH A BELL DRINKER THAN A NIPPLE
DRINKER SUSPENDED ABOVE A TROUGH
G.M. Cronin, K.J. Williams and J.A. Downing - University of Sydney
190
BROILER INTESTINAL HEALTH
MICROBIAL PROFILES OF THE GASTRO-INTESTINAL TRACT OF BROILERS AND ITS
RELATION TO FEED EFFICIENCY
191
L.L. de Lange and P.J.A. Wijtten - De Heus Feeds B.V. The Netherlands
CHARACTERISATION OF GUT BACTERIA ASSOCIATED WITH BROILER PERFORMANCE
ACROSS VARIOUS AUSTRALIAN FEEDING TRIALS
V.A. Torok, K. Ophel-Keller, L.L. Mikkelsen, R. Perez-Maldonado, K. Balding, R.
MacAlpine and R.J. Hughes - SARDI
195
EFFECTS OF PHYTOGENICS AND ZINC BACITRACIN ON PERFORMANCE AND INTESTINAL
HEALTH STATUS OF BROILERS
199
R.M. Gous, T. Steiner and R. Nichol – Biomin Australia
THE ABILITY OF GREEN TEA TO POSITIVELY MODIFY THE GUT MICROFLORA IN BROILER
203
CHICKENS
D.V. Thomas, A.L. Molan and V. Ravindran – Massey University
EFFECTS OF NOVEL FEED ADDITIVES ON GUT HEALTH AND OVERALL PERFORMANCE IN
BIRDS CHALLENGED WITH CLOSTRIDIUM PERFRINGENS
207
M.S. Geier, L.L. Mikkelsen, V.A. Torok, G.E. Allison, C. Olnood, A.Setia, M. Choct,
M. Boulianne and R.J. Hughes - SARDI
FEED ADDITIVES INFLUENCE GOBLET CELL DISTRIBUTION AND VILLUS-CRYPT
ARCHITECTURE IN BROILERS AFTER NECROTIC ENTERITIS CHALLENGE
211
H.M. Golder, M.S. Geier, P.I. Hynd, R.E.A. Forder, M. Boulianne and R.J. Hughes
- University of Adelaide
AUTHOR INDEX
215
Aust. Poult. Sci. Symp. 2010...21
RECOGNITION OF 100TH ANNIVERSARY OF THE FACULTY OF VETERINARY
SCIENCE – UNIVERSITY OF SYDNEY
It is my pleasure, as the Faculty of Veterinary Science’s new Dean, to welcome you to the 21st
Annual Australian Poultry Science Symposium. This symposium promises to continue in the
Poultry Research Foundation’s proud tradition of forging new research directions and
facilitating collaborative links between poultry scientists, producers, researchers and
veterinarians. The symposium will draw together leaders from Australia and our local region
with international experts and provide participants with access to the most current insights into
improving poultry health and production. The symposium topics span the breadth of nutrition,
metabolism, physiology, behaviour, welfare, microbiology, husbandry, toxicology and immune
function in intensive production systems. Presented papers will also address the issues facing
poultry production in developing countries, helping our near neighbours to achieve greater food
security for their people. It is particularly appropriate that the PRF symposium is one of our
first celebrations of our Centenary Year as the PRF is one of the longest established
Foundations of the Faculty and the University. It is also a Foundation that grows in strength
every year, thanks to the dedicated efforts of the Foundation members and supporting staff. We
appreciate your contributions and look forward to stimulating discussions on poultry
production in a time of global uncertainty and to welcoming you to other events in the Faculty
Centenary Program, 2010.
HISTORY OF THE FACULTY
In 1909 the University of Sydney, with the support of the New South Wales Government,
established a Veterinary School and appointed James Douglas Stewart, MRCVS, the Director
and Professor. The school officially opened in 1910 with 16 students enrolled in the 5 year
Bachelor of Veterinary Science. The teaching moved from the basement of the southwest
corner of the Quadrangle (original Fisher library), to the new JD Stewart Building late in 1913.
The school’s development was slow in the first world war, but with the leadership of Professor
Stewart, who remained Dean until 1939, it grew steadily to 25 undergraduates (1928), and over
100 in 1935. The Sydney University Veterinary Faculty was the only one in Australia to
operate continuously from its establishment.
In 1936 the University, in association with the McGarvie Smith Institute, purchased a 160
hectare property at Badgery’s Creek, for Animal Husbandry teaching, and at this time the
degree reverted to a 5 year long course. Building extensions on main campus (1939), were
followed by the temporary Department of Veterinary Pathology and Bacteriology building
(1946) the Veterinary Teaching Hospital (1949). In 1954 the Camden farms were acquired to
provide final year students with animal units for the teaching of husbandry and disease control,
and with a Veterinary Clinic and Hospital, Lecture Theatres and Teaching Laboratories, and a
hall of residence (Nepean Hall).The extensive Hospital and Clinic buildings (Evelyn Williams
Building), an Animal Science building (RMC Gunn Building) were added, the McMaster
building acquired and the Veterinary Science Conference Centre (opened 1998) were erected at
the Camperdown Campus (Sydney).
The Faculty expanded and internationalised its veterinary student intake in 2000 with the
introduction of the new curriculum, which coincided with major changes to the Faculty
structure, removing existing Departments. To meet increased demand, the faculty offered a new
undergraduate degree, the Bachelor of Animal and Veterinary Bioscience in 2005. This 4-year
i
Aust. Poult. Sci. Symp. 2010...21
degree provides a quality education in structure and function of animals, their management and
welfare in an agricultural, para-veterinary, laboratory or wildlife context. We have also
introduced a number of postgraduate diplomas and coursework programs, leading to Master of
Animal Science, Master of Science in Veterinary Science, Master of Veterinary Science,
Master of Veterinary Studies, Master of Veterinary Clinical Studies and Doctor of Philosophy.
The current undergraduate student profile is approximately 80% female, with special entry
programs for indigenous, rural-origin, mature age and disadvantaged students. In the past
decade the Faculty’s staff numbers, research funding, productivity and postgraduate student
training have more than doubled, along with the substantial growth in student numbers to more
than 900 in 2010.
The Wildlife Health and Conservation Centre (2004), General Teaching Block (May 2007) and
the Liz Kernahan Conference Centre (2009) were added to the Camden campus. The Camden
2020 master plan has commenced with current developments in Sample Biobanking facilities,
Surgical Training facilities and student accommodation (duplexes for 70) due for completion in
2010.
Professor JD Stewart,
Founding Dean, 1909-1939
Sir Ian Clunies Ross,
Dean 1940-1946
Anatomy dissection in basement of Quadrangle, 1910
1977)
Old veterinary surgery (demolished
Acknowledgments:
Photos: Veterinary science foundation and faculty history archives, maintained by Professor Paul
Canfield
History: Faculty of Veterinary Science Handbook and Professor Paul Canfield
ii
Aust. Poult. Sci. Symp. 2010...21
HISTORY OF THE POULTRY RESEARCH FOUNDATION
The PRF was one of the original University of Sydney foundations, established in October
1958. It has been a resounding success in its mission to support liaison and research
collaboration between the University and the poultry industry. Its success has come through the
strong partnerships forged among the Foundation members, poultry and stock food industry
leaders and Faculty research staff and students. Together, particularly through the annual
symposium, they have contributed to improvements, particularly in the feeding and husbandry
of chickens, turkeys, ducks and other commercial poultry. The symposium has achieved greater
international recognition and regularly is linked with other leading events in the world poultry
research calendar. The research supported by the Foundation has assisted in refining the
nutrient requirements and optimal diet formulations for poultry production, issues of great
concern as feed is the major cost in production. The early Foundation Chairs, particularly Mr
John Darling, Mr Val Parkinson, Dr Balkar Bains and of course our current President, Ms
Linda Browning, have forged the productive relationships that have been central to the
Foundation’s success.
The Foundation has also had a marked impact on research directions, postgraduate student
training and undergraduate education in the University, through the appointment of outstanding
poultry research scientists and educators. The Directors have provided imagination, energy and
intellectual drive for new research directions and developments, particularly Professor Terry
Robinson, Professor Frank Annison, Professor David Fraser, Professor Tom Scott and Dr Peter
Groves. Not to mention the insight of inaugural directors from the Department of Animal
Husbandry, later to become the Department of Animal Science, Dr. Harold McNary, Associate
Professor Charles Payne and Associate Professor Derick Balnave. These directors have been
supported by a capable and committed team of technical, administrative and academic staff,
who ensure the diverse range of the Foundations’ goals are achieved, and we are most fortunate
to have Dr Wendy Muir, Dr Jeff Downing, Dr Greg Cronin and Dr Peter Selle, Mrs Jo-Ann
Geist, Mrs. Joy Gill and Mrs. Melinda Hayter currently on staff.
The PRF has a bright future, with many opportunities to further improve the health, welfare and
productivity of poultry ahead. Thank you for your diverse contributions and we welcome your
close and productive association with the PRF, Faculty and University.
L to R: Professor Terry Robinson, Mr John Darling, Dr Harold McNary, Associate Professor Charles
Payne 1975 Poultry Science Award,
iii
Aust. Poult. Sci. Symp. 2010...21
Construction of the first layer shed 1959
(Dr. Harold McNary & Mrs.Janet McNary)
Layer shed in operation, early 1960s,
Acknowledgments:
Photos: Poultry Research Foundation
th
History: Preface, 2008 50 Anniversary Seminar (Professor Wayne Bryden)
All my best wishes for a most successful meeting,
Rosanne Taylor
Dean, Faculty of Veterinary Science, 2010
iv
Aust. Poult. Sci. Symp. 2010...21
JOHN L. BARNETT IN MEMORIAM
The tragic loss of Associate Professor John Barnett and his wife Jenny Barnett in the
Victorian bush fires on February 7th 2009 deeply affected all those who knew and loved them.
John was a true gentleman, a brilliant scientist, a world leader in the field of animal welfare
science and above all a true humanitarian.
John’s main area of expertise was stress physiology and its application to the study of domestic
animal welfare. This research over 30 years provided a timely balance on discussions within
science and the livestock industries on welfare methodology and interpretations and this impact
will continue to improve animal welfare methodology in the future. John’s research on poultry
and pigs have also made a critical contribution to our understanding of the welfare risks associated
with confinement housing, highlighting the major risks of confinement that arise from spatial and
social restriction. He worked extensively with the livestock industries in developing welfare
components of livestock industry QA programs and in assisting to achieve improvements in
awareness and practices to safeguard animal welfare standards. His outstanding scientific efforts
have been highly acclaimed nationally and internationally by both science and the livestock
industries and animal welfare science will greatly miss his important contributions.
John was major contributor to the annual Australian Poultry Science Symposia and triennial
Australian Poultry Conventions, and was an active contributor to symposia and meetings of
the WPSA European Federation Working Group 9 on Poultry Welfare and Management. He
led the Australian Poultry Cooperative Research Centre’s Welfare program since 2003. John
was also a major contributor to the Animal Welfare Science Centre’s research and teaching
programs at the University of Melbourne, Monash University and the Victorian Department
of Primary Industries. His wise counsel on matters of science as well as life will be sadly and
irreplaceably missed.
v
Aust. Poult. Sci. Symp. 2010...21
DEVELOPMENTS AND INNNOVATIONS IN BROILER NUTRITION
IN THE NETHERLANDS
L.L. DE LANGE 1
Summary
During the last decades important triggers for innovation have been the reduction of the
output of minerals and nitrogen to the environment and the ban by the EU on the preventive
addition of antibiotic growth promoters (AGP’s) to feeds. A good protein evaluation system
with an accurate estimation of requirements will minimise the output of nitrogen to the
environment, improve intestinal health and performance. AGP’s have a direct effect on the
numbers of bacteria in the gut but also on the composition of the microbial community in the
different segments of the gastro-intestinal tract of broilers. Relative high quantities of DNA
from Lactobacillus Acidophilus in the upper gut of young broilers are related to a bad feed
utilisation. The effect of immune modulators on performance is not always consistent and
depends on source, age and challenge. Future challenges in animal nutrition research are
about animal welfare and how to reduce the output of green house gasses in animal
husbandry to prevent global warming.
I. INTRODUCTION
The Netherlands (NL) is a small country, but big in agriculture. With an agricultural export
volume of about 58.5 billion Euro NL is the world's second largest exporter of agricultural
and food products after the USA. The net trade surplus of the agricultural sector in NL of 23
billion Euro in 2007 is after the US the second largest in the world, and represents 57% of the
total Dutch trade surplus (LEI, 2009). This strong position is built on the leading role of The
Netherlands in primary agricultural production, logistics and agricultural research.
In the seventies and eighties of the last century the livestock industry in The
Netherlands has developed rapidly. The main reasons for this development were the low
transport costs due to the favourable infrastructure and waterways, the high prices of grains
within the European Union (EU) combined with the use of cheap imported alternative feed
ingredients as tapioca from Thailand and maize gluten feed from the US. All this has led to
low feed prices and low feed costs per kg meat, milk or eggs in The Netherlands compared to
other European countries. So the turnover of the feed industry in NL could increase for many
years with 5 – 10 % annually.
In the nineties, the high stocking density close to a dense human population caused
problems with pollution, and led to restrictions for the output of nitrogen, minerals and trace
elements to the environment. In the last decade, the risks of a high stocking density have
become clearer with the outbreak of contagious diseases. The advantage of the import of
cheap raw materials disappeared when the prices of the cereals came down by a changing
policy of the EU. By this reason and the environmental restrictions the feed production went
down from 16.5 million tons at the start of the nineties to a stable volume of about 14 million
tons during the last five years. The production of poultry feeds has been for many years quite
constant on a level of 3.5 million tons a year, disregarding the severe outbreak of avian
influenza in 2003. Of the 3.5 million ton poultry feed about 1.5 million ton is for broilers and
about 2 million ton concerns feed for layers (FEFAC, 2009).
1
De Heus Feeds BV, P.O. Box 396, 6710BJ Ede, The Netherlands
1
Aust. Poult. Sci. Symp. 2010...21
All this made the animal feed industry in NL very competitive, innovative and
expanding abroad. The knowledge about the evaluation of raw materials coming from all
over the world (CVB-tables), about the nutritional requirements of farm animals, about how
to minimise the output of nitrogen and minerals to the environment (phytase is a Dutch
invention) has developed well during the last decades within The Netherlands and more
specific within De Heus Feeds.
The gross domestic production of poultry meat in NL is 700,000 tons per year. The
consumption of broiler meat in The Netherlands comes mainly from breast and processed
meat, while the local consumption of grillers and leg meat is relatively low. The average
consumption of broiler meat in NL is 18.4 kg per capita or 303,000 ton total per year. The
main export markets are Germany and the UK (PVE, 2009). The 44 million broilers in NL
are housed on 700 farms. So the average size of the family owned farms is around 70,000
broilers, taking into account the time that the houses are empty and cleaned. The life weight
at slaughter varies from 1.7 till 3.5 kg with an average of about 2.3 kg. The broiler houses are
well insulated with a good climate control, and cleaned and disinfected after each flock. Feed
costs are about half of all production costs. The costs for housing are almost 10%.
The production of broiler meat in NL is not integrated. Slaughter houses, broiler
farmers, hatcheries and feed mills operate independently, but they make short term contracts
and exchange information. De Heus Feeds has in this highly competitive broiler market a
market share of 25 – 30% of the compound feed.
II. DEVELOPMENTS
In an open competitive market economy is always a good trigger for rapid developments and
innovations. With the decline of prices of cereals in the EU at the nineties adding whole
wheat on the broiler farm has become popular saving processing and transport costs. At the
beginning about 10% whole wheat was used combined with standard feeds without affecting
the performance negatively. Nowadays frequently 20 – 35% whole wheat is used together
with concentrates rich in protein.
Environmental problems induced the development of phytase by Gist Brocades and
Dutch research institutes. The use of this enzyme has become widespread all over the world
when the use became economic by replacing expensive mineral phosphorous and by the
beneficial side-effects of phytase by reducing the anti nutritional effects of phytate.
The announcement of the ban of Antibiotic Growth Promoters (AGP’s) in feeds
starting in 2006 by the EU has given an enormous boost to research on intestinal health and
how to replace these feed additives. Many alternatives for AGP’s have been tested without
questioning the mode of action of AGP’s. More recently, efforts have been put to reduce the
therapeutic use of antibiotics at farm animals on prescription by veterinarians. Especially the
increasing concern about life stock farmers carrying resistant bacteria in their body is an
important motive to reduce also the therapeutic use of antibiotics.
Another concern is about animal welfare starting in the UK more than 10 years ago.
In The Netherlands a one issue political party on animal welfare is even represented in the
parliament. EU regulations for animal welfare especially about housing are nowadays
common practise. A new EU regulation starting July 2010 about stocking density related to
mortality and foot pad dermatitis will affect the broiler industry. If the norms for mortality or
foot pad dermatitis are exceeded the maximum stocking density has to be reduced (EU,
2007).
Also retailers in The Netherlands try to profile themselves recently with themes as
sustainability and animal welfare. Special niche markets are being developed with extra
added value and more attention for animal welfare.
2
Aust. Poult. Sci. Symp. 2010...21
III. INNOVATIONS
(a) Protein evaluation
In The Netherlands the protein requirements for poultry are usually based on the faecal
digestibility of amino acids (AA). Minimum requirements are set for the essential amino
acids as lysine, methionine + cysteine and hreonine. The other essential amino acids are
normally not limiting. If some nonessential AA are not supplied sufficiently these can be
synthesized from non limiting essential and other nonessential AA. To ensure that the
requirements for all AA are provided, one can formulate requirements for both essential
amino acids and crude protein (NRC, 1994). An indirect way to ensure the supply of
nonessential AA is by maximising the use of synthetic AA. At De Heus we prefer to
formulate diets with protein requirements for essential AA and the sum of all digestible AA
(De Lange et al, 2003). Doing so, also the surplus of non limiting essential AA is utilised to
meet the requirements for nonessential AA. By this approach and by using well digestible
protein sources, the amount of nitrogen at the end of the intestinal tract is minimised to
prevent proteolytic fermentation. The fermentation of protein can increase the growth of
pathogenic bacteria such as sulphite reducing clostridia and the formation of toxic
compounds as ammonia and phenol as is demonstrated after predigestion with in vitro
research in the TNO Intestinal Model (TIM-2) to simulate the function of the hindgut (Table
1).
Table 1. End products proteolytic fermentation TIM-2
Feed
A
B
Feathermeal (% in feed)
0.0
0.0
Digestibility protein (%)
92.5
91.5
Fermentable protein (g/kg residue)
73
77
Cumulative production of:
Ammonia (mmol)
9.5
13.8
Phenol (µmol)
0.0
0.0
10
5.1
5.8
Sulphite reducing Clostridia ( log cfu/ml)
C
1.5
89.5
96
D
3.0
82.2
138
13.1
8.9
6.3
15.4
14.0
6.8
To avoid proteolytic fermentation one can minimise the quantity of fermentable
protein (Cone, 2005), closely related to indigestible protein or the difference between ileal
digestible protein as described by Bryden et al (2009) and faecal digestible protein.
The objective of a correct feed evaluation is to predict animal performance accurately
with different compositions of the feed. The combination of the minimum for digestible AA
and a maximum for indigestible crude protein gives a better prediction of animal performance
than crude protein (De Lange et al, 2003).
The protein levels in feed depend on age/weight, breed, intake, gain and economy.
With the modern broilers there is a good relation between protein level in the feed and animal
performance as is shown in trial VK-062, performed by De Heus (Figure 1 and 2).
3
Aust. Poult. Sci. Symp. 2010...21
2300
Life weight at day 35 (g/brolier)
R2 = 0.870
2200
Cobb 500
Ross 308
2100
R2 = 0.921
2000
1900
Very low
Low
Medium low
Medium high
Protein level in feed
High
Very high
Figure 1. Relation between protein in feed and life weight at day 35 (VK-062)
FCR at 35 days at 2 kg LW (kg/kg)
1.70
Cobb 500
1.65
Ross 308
1.60
1.55
R2 = 0.894
2
R = 0.968
1.50
1.45
Very low
Low
Medium low
Medium high
Protein level
High
Very high
Figure 2. Feed Conversion Rate per cross breed and per protein level at day 35
at 2 kg Life Weight (correction 0.03 per 100 g LW; VK-062)
After correction for protein requirements for maintenance, the protein conversion
expressed as protein requirements per 100 g growth is equal for the main cross breeds during
the grower or finisher period (Figure 3). Taking into account the differences in intake and
growth, the protein requirements expressed per kg feed are different between the main
breeds. There seems to be a continuous improvement in protein efficiency in time when the
protein requirements are expressed per 100 g daily gain.
4
Aust. Poult. Sci. Symp. 2010...21
94
Average daily gain (g/day)
89
y = -0.036x2 + 3.406x + 21.087
R2 = 0.961
84
79
74
69
64
15
17
19
21
23
25
Protein intake (g DAA/day)
27
29
31
Figure 3. Relation daily protein intake and daily gain (VK-062)
(b) Intestinal health
New molecular DNA based techniques like Terminal Restriction Fragment Length
Polymorphism (T-RFLP), as described by Lu et al (1997) reveal the complex microbial
societies in many biotopes as the broilers gut. In the ileum of broilers mainly Lactobacilli are
dominant, while in the cecum Clostridiaceae related species are abundant (Lu et al, 2003).
AGP’s reduce the overall numbers and the numbers of species of gut bacteria in pigs and
poultry (Jensen, 1998; Gaskins et al, 2002). Bacteria in the gut may be considered as
parasites or commensals, dining at the same table as the host and consuming nutrients.
Bacteria like Enterococcus faecium and Clostridium perfringens but also Lactobacillus
species hydrolyse bile salts and cause impaired digestion of fat (Knarreborg et al, 2002).
Bacteria also trigger the immune system as is shown in germ-free animals, which have a
poorly developed immune system. Triggering the immune system leads to an acute phase
response with loss of appetite and catabolism of muscle tissue (Gruys et al, 2006). Niewold
(2007) hypothesizes that the growth promoting effect of AGP’s relies on an antiinflammatory effect by inhibiting the production and secretion of cytokines by intestinal
inflammatory cells and so reducing the acute phase reaction.
In a cooperative research by Provimi and De Heus Feeds, funded by the Dutch
Ministry of Economic affairs, is shown that AGP’s and feed composition have an effect on
the microbial composition in the gut, which is related to feed efficiency. Relative high
quantities of DNA from Lactobacillus Acidophilus in the upper gut of young broilers are
related to a bad feed utilisation, and are affected by feed and AGP’s (De Lange and Wijtten,
2008). Davis et al (2009) showed a relation between the relative numbers of L. Acidophilus in
the gut of pigs and a systemic immune response by measuring T-cells in the peripheral blood.
The alternative of De Heus Feeds for the use of AGP’s is called Nutribiotics®. This
concept is based on a high digestibility of protein to prevent the formation of toxic
metabolites like ammonia and phenol and to prevent increased numbers of sulphite reducing
Clostridia, on natural antimicrobial components to reduce bacterial growth, on a feed form
with coarse particles to lower the pH and to increase the retention time of feed in the gizzard,
and on a low viscosity of the intestinal content to facilitate digestion. After the ban De Heus
has succeeded in continuing the improvement of broiler performance in practise with a FCR
of 1.70 – 1.75 at a life weight of 2.3 kg.
5
Aust. Poult. Sci. Symp. 2010...21
The next challenge in research and application into practise is the modulation of the
immune system of the broiler chicken. The best way of immune modulation is most probably
by stimulating the development of the immune system at a young age and by reducing acute
phase reactions at an older age when birds have to grow efficiently. Also modulating the
immune reaction to an anti-inflammatory Th2 response instead of a pro-inflammatory Th1
response can help to improve feed and protein conversion. Modulating immunity of young
chickens starts with the parent stock. A well developed innate immune system helps in
initiating the adaptive immune response (Goddeeris, 2005). Feeding BioMos®, a yeast cell
wall (YCW) product to breeder animals changes the ratio yolk/egg white in breeding eggs
and increases the quantity of specific immune globulins in blood against New Castle Disease
after vaccination (De Lange, 2007).
Table 2. Effect of YCW products in breeder diet (VK-078)
P-values
Breeder diet
2
1
Control
b
BioMos )
2080
Weight (g)
2117
FCR
1.597
1.592
FCR 2 kg LW
1.562
1.568
ab
Fibosel )
2068
SEM Br. diet Br.*Chall.
a
14
0.06
0.68
1.578
0.007
0.11
0.05
1.557
0.009
0.65
0.28
1
) BioMos®: YCW product of Alltech; added 1 kg per ton feed
) Fibosel®: purified YCW product of Lallemand; added 200 g per ton feed
a,b
Means within a row not sharing a common superscript differ significantly (P<0.05)
2
In a De Heus experiment (VK-078) adding Fibosel® to the breeder diet reduces the
live weight of the off spring at day 35 compared to the control group (P < 0.05), and tends to
improve the not corrected FCR (P < 0.10) compared to the control group (Table 2). Also the
interaction between the addition of YCW based products to breeder diets and a challenge
with a viscous diet during the grower phase on the FCR tends to be significant (P < 0.10)
(Table 2 and 3).
Table 3. Interaction between YCW in breeder diet and challenge grower diet (VK-078)
Non challenged
FCR
FCR relative
Challenged
Contr BioMos Fibosel
Contr
1.569
1.625
1.602
1.582
100.0%
98.6%
97.4%
1.582
1.573
100.0% 100.8% 100.3%
BioMos Fibosel
The beneficial effect of an investment in the immune system on a young age seems to depend
on a challenge later in life.
6
Aust. Poult. Sci. Symp. 2010...21
Table 4. Effect of YCW products in starter diet (VK-078)
P-values
Starter diet
Control
Weight (g)
2093
FCR
1.595
FCR 2 kg LW
1.567
1
2
BioMos )
2066
1.595
ab
1.575
SEM Br. diet Br.*Chall.
Fibosel )
2107
1.577
b
1.545
a
14
0.14
0.29
0.007
0.15
0.68
0.009
0.08
0.82
1
) BioMos®: YCW product of Alltech; added 1 kg per ton feed
) Fibosel®: purified YCW product of Lallemand; added 100 g per ton feed
a,b
Means within a row not sharing a common superscript differ significantly (P<0.05)
2
Adding Fibosel to a starter diet improves the corrected FCR at 2 kg live weight
compared to the BioMos group significantly (P < 0.05) and the difference with the control
group tends to be significant (P < 0.10) (Table 4).
IV. FUTURE CHALLENGES
The nearby challenge in broiler nutrition science is to improve animal welfare by reducing
mortality and foot pad lesions. The most obvious way to lower mortality is improving health
and robustness of the one day old chickens, starting at breeder farms and hatcheries, but also
at the start on the broiler farms. Reducing feed intake and average daily gain might be also
part of a concept to lower mortality and sudden death. Less foot pad lesions will be pursued
by nutrition and improving litter quality by nutritional and other means.
The best way to respect the environment at the production of broiler meat is by
reducing the output of nitrogen, minerals and trace elements, by improving feed utilisation
and by maximising the use of by-products that are less attractive for direct human
consumption.
REFERENCES
Bryden WL, Li X, Ravindran G, Hew LI, Ravindran V (2009) Ileal Digestible Amino Acid
Values in Feedstuffs for Poultry RIRDC. Canberra, ACT.
Cone JW, Jongbloed AW, Van Gelder AH and De Lange L (2005) Animal Feed Science and
Technology 123-124, 463-472
Davis E, Rehberger J, King M, Brown DC, Maxwell CV, and Rehberger T (2009)
Proceedings, 11th International Symposium on Digestive Physiology in Pigs
De Lange LLM, Rombouts C and Oude Elferink G (2003) World’s Poultry Science Journal,
59, 447-457.
De Lange LLM, Kocher A and Beeks W (2007) Proceedings 27th Western Nutrition
Conference. Alltech symposium, Lexington, Kentucky, USA
De Lange LLM and Wijtten PJA (2008) Proceedings 33rd Meeting of Dutch speaking
nutrition researchers (NVO) Wageningen, The Netherlands 85-86
EU (2007) COUNCIL DIRECTIVE 2007/43/EC of 28 June 2007 laying down minimum
rules for the protection of chickens kept for meat production.
http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2007:182:0019:0028:EN:PDF
FEFAC (2009) Compound Feed Production EU (1989-2008).
http://www.fefac.org/file.xls?FileID=15062
Gaskins HR, Collier CT, Anderson DB (2002) Animal Biotechnology 13, 29-42
Goddeeris BM (2005) Conference Antimicrobial Growth Promoters: Worldwide Ban on the
Horizon, Noordwijk aan Zee, The Netherlands
7
Aust. Poult. Sci. Symp. 2010...21
Gruys E, Toussaint MJM, Niewold TA, Koopmans SJ, Van Dijk E, Meloen RH (2006) Acta
Histochemistry 108, 229-232
Jensen BB (1998) Journal Animal Feed Science 7, 45-64
Knarreborg A, Engberg RM, Jensen SK, Jensen BB (2002) Applied and Environmental
Microbiology 68, 6425-6428
LEI (2009) Agri-monitor 2008.
http://www.lei.dlo.nl/nl/content/agri-monitor/pdf/Agrarische%20handel.pdf
Liu WT, MARSH TL, CHENG H, FORNEY LJ (1997) Applied and Environmental
Microbiology 63, 4516-4522.
LU J, IDRIS U, HARMON B, HOFACRE C, MAURER, JJ, LEE MD (2003) Applied and
Environmental Microbiology 69, 6816-6824.
Niewold TA (2007) Poultry Science 86, 605-609
NRC (1994) Nutrient requirements of Poultry. Ninth Revised Edition edited by National
Academy Press, Washington, D.C. USA
PVE (2009) Statistics Livestock, meat and eggs 2008.
http://www.pve.nl/wdocs/dbedrijfsnet/up1/ZcfpdjdIE_421198PVVfolderEN.pdf
8
Aust. Poult. Sci. Symp. 2010...21
CHALLENGING CURRENT POULTRY FEEDING DOGMAS BY FEED INTAKE
RESTRICTION AND THE USE OF COARSE FEED INGREDIENTS
B.SVIHUS 1
Summary
In Europe, there is a trend towards a coarser structure of the diet. In the broiler industry, there
is also growing interest for restricted feeding. A coarser structure of the diet reduces the cost
of feed production, and has also been linked to an improved feed utilization through a better
functioning gizzard. A slightly restricted feeding has also been observed to result in
improvement in feed efficiency, and this has been shown to be caused by an altered growth
curve and composition of growth. Although feed overconsumption is observed in individual
broiler chickens under certain conditions, the bird’s ability to store large quantities in the crop
and gizzard makes it uncertain whether feed restriction may be a suitable tool to reduce this
problem. However, a coarser structure may reduce the problem of feed overconsumption
through a better functioning gizzard.
I. INTRODUCTION
Currently, the dominating global principle of poultry feeding is to provide poultry ad libitum
with one diet containing finely ground low-fibre ingredients. The advantage of this feeding
practice is obvious; it allows for a simple management, and it assures a homogenous feed
which is easily available for digestion by the bird. However, there is emerging evidence for
that this feeding principle does not maximise performance and profits. In this paper, the
paradigm that birds should be ad libitum fed a diet containing finely ground low-fibre
ingredients will be challenged, using the broiler chicken as an example.
The practice of fine grinding has been challenged in Europe during the past 20 years,
partly due to data which indicate that coarser grinding or the use of whole cereals saves cost
and in some cases has positive effects on bird performance (Preston et al., 2000; Hetland et
al., 2002: Plavnik et al., 2002; Svihus et al., 2004; Ravindran et al., 2006; Gabriel et al.,
2008). In addition, similar beneficial effects have been observed when coarse fibre
components such as oat hulls or wood shavings are added to the diet (Hetland et al., 2003a;
Amerah et al., 2009). The beneficial effect of coarse ingredients has been linked to a more
developed gizzard that ensures a complete grinding and a well regulated feed flow and
digestive juice secretion.
Although ad libitum provision of feeds to broilers remains the dominating practice,
different feed restriction programmes appear to be on the rise. In Norway for example, it has
now become common practice with a slight restriction on feed availability. Although the
restriction is so moderate that the growth curve is not significantly altered, a slightly
improved feed efficiency and a reduced mortality is commonly observed, favouring increased
use of this feeding method. A common management practice is to restrict feed intake slightly
by letting the birds empty the feeders twice a day, and to provide feed according to a
predetermined feeding plan. The beneficial effect of this type of feed restriction could simply
be a management benefit due to the fact that feeders are emptied regularly and thus that fines
are not accumulating. However, there are indications for that some birds under ad libitum
1
Norwegian University of Life Sciences, P.O. Box 5003, N-1432 Aas, Norway.
[email protected]
9
Aust. Poult. Sci. Symp. 2010...21
feeding may develop a habit of feed over-consumption (Svihus & Hetland, 2001), which if
correct calls for restrictions on feed availability. Also, there is overwhelming evidence for
that intermittent feeding by the use of lighting control improves feed efficiency compared to
ad libitum feeding (Buyse et al., 1996).
Since birds under restricted feeding depend on temporary storage of feed in the
anterior digestive tract, the storage capacity of this digestive segment is of importance. It has
been shown that broiler chickens use both the crop and the proventriculus/gizzard as storage
organs for food when adapted to long periods of food deprivation (Buyse et al., 1993). Barash
et al. (1993) observed a significant increase in weight and feed-holding capacity of both crop
and gizzard when chicks were fed meals one or two times per day instead of ad libitum. It is
well known that structural components in the feed result in a larger gizzard, which will also
be able to hold larger quantities of feed. It is therefore possible that an interaction between
feed restriction and the structure of the diet exists.
II. STRUCTURAL COMPONENTS IN THE DIET
It is a well established fact that the digestive tract adapts to changes in diet composition.
Studies of wild birds with large variations in their diet throughout the year for example, show
that both the small intestine and the caeca fluctuate in size (Klasing, 1998). Particularly the
gizzard is known to respond rapidly to diet changes, something noted amongst others by
Charles Darwin during his studies of pigeons (Farner, 1960). A rapid and conspicuous
enlargement in size of the gizzard is observed when structural components such as hulls,
wood shavings or large cereal particles are included in the diet, and the predominant
hypothesis is that the beneficial effects sometimes observed is mechanistically linked to
gizzard development.
The increase of size of the gizzard is a logical consequence of an increased need for
particle size reduction, as the increased grinding activity of the gizzard increases the size of
the two pairs of gizzard muscles. For example, Hetland et al. (2003a) observed a 50% heavier
gizzard when whole cereals and oat hulls were included in the diet for broiler chickens.
Coarse feed particles need to be ground to a certain critical size before they can leave the
gizzard (Clemens et al., 1975; Moore, 1999), causing the volume of gizzard contents to
increase when diets with whole cereals or insoluble fibre are fed (Hetland et al., 2003a,b).
Thus, structural components do not only increase the size of the gizzard, but also causes an
increased volume of gizzard contents in birds fed diets with whole grains or other structural
components (Hetland et al., 2003a,b).
An increased amount of finely ground particles in the duodenum for diets with
structural components indicates that one potential beneficial effect of structural components
is smaller particles entering the small intestine and thus a more efficient and complete
digestion (Hetland et al., 2002). This corresponds to the increased starch digestibility
sometimes observed when structural components have been added (Svihus & Hetland, 2001;
Svihus et al., 2004). However, the increased amylase activity and bile acid concentration
observed by Svihus et al. (2004) may indicate that an increased secretory activity may also be
part of the cause for improvements in nutritive value associated with structural components.
This hypothesis is also supported by the fact that amylase activity in jejunum and starch
digestibility in anterior, median and posterior ileum for individual birds had a correlation of
0.56, 0.54 and 0.47, respectively. The cause for this increased secretory activity remains
unclear, but may be associated with a stimulation of pancreatic secretion caused by an
increase in gizzard activity. Hetland et al. (2003a) found a significant increase in amylase
activity and bile acid secretion when gizzard activity was stimulated by oat hulls.
10
Aust. Poult. Sci. Symp. 2010...21
The major stimuli causing an increased enzyme secretion by the pancreas are the
vagus nerve and cholecystokinin (CCK). Cholecystokinin is mainly produced in the pyloric
region of birds (Denbow, 2000), and acts together with the vagus nerve to stimulate
pancreatic enzyme secretion (Li and Owyang, 1993). Thus, a stimulating effect of an
increased gizzard activity on enzyme secretions may be mediated through increased CCK
release in the pyloric region of the gizzard.
Since an increased volume of the gizzard and thus an increased amount of feeds is
found in birds fed structural components (Amerah et al., 2009; Svihus et al., 2004), the
resulting increased retention time may also affect nutrient digestibility. It has been shown that
pH of gizzard content decreases when the diet contains structural components (Gabriel et al.,
2003). The cause for this could be the increased retention time as a consequence of increased
gizzard volume, which would allow for more time for hydrochloric acid secretion. Although
pepsin production has not been measured directly, a common secretory pathway for
hydrochloric acid and pepsinogen make an assumption of increased pepsin secretion
probable. Thus, another possible explanation for improvements in digestibility with
structural components could be a more complete digestion in the stomach. Increased retention
time in the gizzard may also potentially improve efficacy of exogenous enzymes added to the
diet. Although this surprisingly enough appears to not have been studied, it is often postulated
that an exogenous enzyme such as phytase exerts its main function in the gizzard due to the
favourable pH there (Garrett et al., 2004). Thus, it is a reasonable hypothesis that a more
developed gizzard as a consequence of structural components may improve efficacy of this
particular enzyme. Since pH optimum for most phytases are in the area 4 to 5 however, the
reduced pH with structural components may counteract the effect of increased retention time.
These important questions call for being addressed in future experiments.
A coarser grinding will save considerable energy and time in the grinding process.
Reece et al. (1986) stated that energy cost for hammer grinding of maize could be reduced
with 27 % by increasing the sieve size from 4.76 to 6.35 mm. Thus, the reduced production
costs and the improvements in performance are major motivations for a coarser structure of
the diet.
III. OVERCONSUMPTION AND FEED INTAKE RESTRICTION
Birds eat and swallow food without prior chewing. Although secretion of saliva do occur, the
saliva do not contain amylase in the chicken and turkey, and the time exposed to saliva in the
mouth is short (Duke, 1986). Thus, intact food particles enter the crop, or, if the gizzard is
empty, pass directly to the proventriculus and gizzard. The crop may store large amounts of
food for a considerable amount of time. Svihus et al. (2002) found up to 30 g feed in the crop
30 minutes after feeding starved broiler chickens weighing between 1100 and 1400 g, with
half this feed still remaining in the crop after 3 hours. In a recent experiment carried out in
cooperation with the Poultry CRC at University of New England, it was shown that crop
contents are rapidly moistened (Figure 1). Although no digestive enzymes are secreted and
no grinding occurs, feed may be softened and microbial and exogenously added enzymes
may be activated. Denstadli et al. (2006) showed that at conditions similar to those in the crop
(45 % moisture, 45 °C and a pH of 4.7), added phytase was able to degrade 86 % of the
inositol 6-phosphates in the feed within 45 minutes of incubation. The effect of crop retention
on the efficacy of exogenous enzymes is currently under investigation at our lab.
11
Aust. Poult. Sci. Symp. 2010...21
Figure 1.
Dry matter concentration of contents of crop and gizzard at different times
after feeding.
The main body of the gizzard comprises two thick, opposed lateral muscles and two thin
anterior and posterior muscles. A thick layer of glyco-proteins that is hardened by the low pH
covers the inside of the gizzard. The koilin layer is composed of a combination of rod-like
and granular secretions from gizzard glands, and is continuously renewed by these glands as
it is worn. Some birds, however, are known to shed the whole koilin layer at intervals
(McLelland, 1979). None of the domesticated bird species have been reported to have such
behaviour, but unpublished observations of gizzards lacking koilin layer at our lab indicate
that this may happen in turkeys. The rod-like hardened glycoprotein will stand out from the
remaining glyco-protein matrix and will give the surface the characteristic sand-paper like
appearance (Hill, 1971). The thick muscles grind material by contractions that rub the
material against the koilin layer on the inside of the gizzard. The small muscles move
material between contractions of the large muscles.
Own research has led to the hypothesis that over-consumption of feeds may
sometimes occur in broiler chickens, resulting in impaired nutrient utilization due to a too
rapid feed passage through the small intestine. The first indication of this was in an
experiment where it was observed that caged broiler chickens fed pelleted wheat-based diets
showed improved nutrient utilization and smaller individual variation when feed intake was
reduced by crushing the pellet (Svihus & Hetland, 2001). As reported in this forum earlier,
plotting of feed intake for individual birds against AME frequently results in a slight inverse
relationship; when feed intake increases, AME is reduced (Svihus, 2003). In the recent
experiment mentioned above, this was very clearly observed (Figure 2). In this experiment,
four out of ten ad libitum fed birds on the finely ground pelleted wheat diet showed signs of
being feed overconsumers, characterised by a normal weight gain, a higher than average feed
intake and an AME value below 10.3.
12
Aust. Poult. Sci. Symp. 2010...21
Figure 2.
Correlation between feed intake (horizontal axis, g) and AME (vertical axis,
MJ/kg) for 40 individually caged broiler chickens receiving a wheat-based
diet.
An equal number of intermittently fed birds with feed available only during 4 one hour
feedings and one two hour feeding during the day exhibited the same signs of feed
overconsumption. This indicates that intermittent feeding is not effective in prohibiting birds
from feed overconsumption. Despite the very strong restrictions on feed availability through
the five feeding times and only 6 hours with feed available compared to 18 for the ad libitum
fed birds, the intermittently fed birds were able to adapt quite quickly to intermittent feeding,
as indicated by the similar weight gain as ad libitum feeding after an adaptation period and no
significant reduction in bird weight at the termination of the experiment. Obviously, this is
due to an extensive use of the crop as an intermediary storage organ for feed. For example, on
day 21 when the average daily feed consumption for all birds were 103 g, four out of 20
intermittently fed birds consumed between 28 and 30 g during one or several of the one-hour
feeding bouts, and 12 out of 20 consumed more than 20 g. The fact that more than two-thirds
of the feed were consumed during the first twenty minutes of the hour and that very little was
consumed during the last twenty minutes of the hour also illustrates the ability of broiler
chickens to consume large quantities of feed during a short time. This adaptability allows the
intermittently fed birds to store large quantities of feed in the crop and gizzard, which can
then be relied upon when feed is not available. Thus, it is possible that this mechanism
hindered any prohibitory effect of the meal feeding on feed overconsumption.
The question may arise on the causes for over-consumption among individual birds.
Peter Siegel was among the first to indicate that intensive breeding for weight gain had
resulted in broilers that were capable of over-consuming feed due to disturbance of the
appetite control centre in the brain (Siegel and Dunnington, 1987). Lacy et al. (1985) showed
that broilers, as opposed to layers, did not respond to intra-hepatic glucose infusions with
reduced feed intake. This indicates that even peripheral appetite control centres have been
disturbed in broilers. It can thus be postulated that modern breeds of broilers may overconsume feeds, with a resulting impairment in nutrition utilization for some diets. Although
Buyse et al. (1996) found improved feed utilization with intermittent feeding, it was
hypothesized that a more concave growth curve for intermittently fed birds which would save
maintenance cost of birds at an early age, was the main explanation, together with an altered
growth composition. The lack of any effect of intermittent feeding in our experiment
mentioned above indicates that compensation mechanisms through intermediate storage in
the crop and gizzard allow even restrictedly fed birds to overconsume feeds. It is therefore
possible that the main beneficial effects of restricted feed availability are an altered growth
13
Aust. Poult. Sci. Symp. 2010...21
curve, possibly together with beneficial effects associated with the altered management, such
as less spillage and less accumulation of feed in the feeders. However, feed restriction
through making the feed less easy to consume appears to be able to hinder feed
overconsumption and improve nutrient availability. In the experiment of Svihus & Hetland
(2001), grinding the pellet to mash resulted in a significant improvement in starch
digestibility which was due to elimination of birds with extremely low ileal starch
digestibility.
IV. DIET STRUCTURE AND FEED OVER CONSUMPTION
Results from our lab indicate that there may be an interaction between gizzard function and
overconsumption of feeds. When feeds have more structure, either through the use of whole
cereals or through addition of large fibre particles (oat hulls or wood shavings) into the diet,
an improvement in starch digestibility and a reduced variation between birds is commonly
observed (Svihus & Hetland, 2001; Hetland et al., 2002; Rogel et al., 1987). This was also
observed in the previously mentioned experiment, where variation in feed/gain, AME and
starch digestibility was strongly reduced for caged Cobb broilers when whole wheat was
added to the diet (Table 1).
Table 1.
Broiler performance (16 to 25 days of age) and nutrient availability1
Ad libitum
Ground
Whole
wheat
wheat
Weight gain, g:
Mean
Coefficient of variation
Range
Feed intake, g:
Mean2
Coefficient of variation
Range
Gain/feed:
Mean3
Coefficient of variation
Range
AMEn, MJ/kg:
Mean4
Coefficient of variation
Range
Faecal starch digestibility:
Mean5
Coefficient of variation
Range
Restricted
Ground
Whole
wheat
wheat
543
0.11
443 – 627
527
0.19
369 – 667
505
0.19
344 – 629
523
0.14
407 – 644
1022a
0.17
783 – 1347
846b
0.19
636 – 1100
890ab
0.27
564 – 1202
795b
0.15
622 – 1016
0.54c
0.16
0.40 – 0.65
0.62ab
0.05
0.58 – 0.67
0.58bc
0.10
0.48 – 0.67
0.66a
0.07
0.59 – 0.72
10.8b
0.15
8.1 – 12.3
12.6a
0.06
10.7 – 13.5
10.9b
0.18
8.0 – 13.1
12.6a
0.07
10.2 – 13.2
0.81b
0.18
0.59 – 0.94
0.95a
0.06
0.81 – 0.99
0.82b
0.18
0.56 – 0.97
0.95a
0.07
0.75 – 0.98
1
ANOVA main effects or interaction effects were non-significant unless mentioned. 2P-value for
wheat structure and feeding system were 0.0213 and 0.1111, respectively. 3P-value for wheat structure
and feeding system were <.0001 and 0.0379, respectively.4P-value for wheat structure was 0.0003.
5
P-value for wheat structure was 0.0004.abcMeans within a row not sharing a common superscript
differ at P<0.05.
14
Aust. Poult. Sci. Symp. 2010...21
Our hypothesis is that an active and well-developed gizzard will function as a food
intake regulator that will assure that the bird does not over-consume feeds. This due to the
active retention of large particles until they are broken down, which results in a larger filling
of the gizzard and thus less room for more feed to be consumed. The hypothesis is that
structural components in the diet will stimulate development of the gizzard, and that this will
result in a better feed flow regulation such that feed overconsumption is avoided. Such a
mechanism may also explain why the effect was so large in the current experiment, as birds
were raised on wire floor with no access to litter material. Previous results indicate that birds
eat litter to compensate for lack of structure in the diet (Hetland et al., 2004; Hetland et al.,
2005).
V. CONCLUSION
The move towards a coarser structure of the feeds in Europe seems justified due to the
reduced feed processing costs and the improvements in performance. Although restricted
feeding has been documented to result in improved feed efficiency, the mechanisms for this
are still unclear, and the value of this practice compared to other changes such as a coarser
structure is thus uncertain. Investigations are currently underway at our lab to substantiate the
feed overconsumption hypothesis further.
REFERENCES
Amerah AM, Ranvindran V, Lentle RG (2009) British Poultry Science 50, 366-375.
Barash I, Nitsan Z, Nir I. (1993) British Poultry Science 34, 35-42.
Buyse J, Simons PCM, Boshouwers FMG, Decuypere E (1996) World's Poultry Science
Journal 52, 121-130.
Buyse J, Adelsohn DS, Decuypere E, Scanes CG (1993) British Poultry Science 34, 699-709.
Clemens ET, Stevens CE, Southworth M (1975) Journal of Nutrition 105, 1341-1350.
Denbow DM (2000) In: Whittow (Ed), Sturkie’s Avian Physiology. New York, Academic
Press.
Denstadli V, Vestre R, Svihus B, Skrede A, Storebakken T (2006) Journal of Agricultural
and Food Chemistry 54, 5887-5893.
Duke GE (1986) In: Sturkie PD (Ed), Avian Physiology. Springer-Verlag, New York, USA.
Farner, DS (1960) In: Marshall, AJ (Ed), Biology and comparative physiology of birds.
Volume 1. New York, Academic Press.
Gabriel I, Mallet S, Leconte M, Travel A, Lalles JP (2008) Animal Feed Science and
Technology 142, 144-162.
Gabriel I, Mallet S, Leconte M (2003) British Poultry Science 44, 283-290.
Garrett JB, Kretz KA, O’Donoghue E, Kerovuo J, Kim W, Barton NR, Hazlewood GP, Short
JM, Robertson DE, Gray KA (2004) Applied and Environmental Microbiology 70,
3041-3046.
Hetland H, Svihus B, Olaisen V (2002) British Poultry Science 43, 416-423.
Hetland H, Svihus B, Krogdahl Å (2003a) British Poultry Science 44, 275-282.
Hetland H, Svihus B, Lervik S, Moe R (2003b) Acta Agriculturae Scandinavica 53, 92-100.
Hetland H, Choct M, Svihus B (2004) World’s Poultry Science Journal 60, 413-420.
Hetland H, Svihus B, Choct M (2005) Journal of Applied Poultry Research 14, 38-46.
Hill KJ (1971) In: Bell, D.J., Freeman, B.M. (Eds), Physiology and biochemistry of the
domestic fowl. Volume 1. London, Academic press.
Klasing KC (1998) Comparative avian nutrition. Wallingford, UK, CAB International.
Lacy MP, Van Krey HP, Skewes PA, Denbow DM (1985) Poultry Science 64, 751-756.
15
Aust. Poult. Sci. Symp. 2010...21
Li Y, Owyang C (1993) Journal of Clinical Investigation 92, 418-424.
McLelland J (1979) In: King AS, McLelland J (Eds), Form and function in birds. London,
Academic press.
Moore SJ (1999) Australian Journal of Zoology 47, 625-632.
Plavnik I, Macovsky B, Sklan D (2002) Animal Feed Science and Technology 96, 229-236.
Preston CM, Mccracken KJ, McAllister A (2000) British Poultry Science 41, 324-331.
Ravindran V, Wu YB, Thomas DG, Morel PCH (2006) Australian Journal of Agricultural
Research 57, 21-26.
Reece FN, Lott BD, Deaton JW (1986) Poultry Science 65, 636-641.
Rogel AM, Balnave D, Bryden WL, Annison EF (1987) Australian Journal of Agricultural
Research 38, 629-637.
Siegel PB, Dunnington EA (1987). CRC Critical Reviews in Poultry Biology 1, 1-24.
Svihus B (2003) Proceedings of the Australian Poultry Science Symposium 15, 17-25.
Svihus B, Hetland H (2001) British Poultry Science 42, 633-637.
Svihus B, Hetland H, Choct M, Sundby F (2002) British Poultry Science 43, 662-668.
Svihus B, Juvik E, Hetland H, Krogdahl Å. (2004) British Poultry Science 45, 1-6.
16
Aust. Poult. Sci. Symp. 2010...21
CHALLENGES TO EUROPEAN POULTRY PRODUCTION
J.D. VAN DER KLIS 1
Summary
European poultry production is regulated by many EU-Directives and national regulations that
focus on food safety, animal welfare and environmental issues. In this paper animal welfare
and control of environmental pollution are dealt with, as EU-27 average and differences
between EU-27 countries. Implementation of the EU-Directives to improve animal welfare
can have large consequences on the competitiveness of poultry production in Europe. Control
of (early) mortality and wet litter of broilers are prerequisites for an economically feasible
broiler production in countries where they are directly linked to stocking density in the near
future. Moreover, direct costs for impaired intestinal health are substantial. Alternatives to
antimicrobial growth promoters based on their potential non-antibiotic anti-inflammatory
response have added value to broilers and floor-housed laying hens with impaired intestinal
health. Dietary dilution of broiler breeder diets might help to reduce chronic hunger stress and
improve broiler vitality especially in young broiler breeders.
I. INTRODUCTION
European poultry production is regulated by many EU-Directives and additional national
regulations that focus on food safety, animal welfare and environmental issues. On the one
hand legislation requires investments in e.g. best available techniques to reduce environmental
pollution, or alternative housing systems, but on the other hand it limits farm size (e.g. via egg
production rights in the Netherlands) and stocking density, specifying a minimum space
allowance for layers and broilers. Especially the EU-Directives to improve animal welfare
will have a large impact on the European poultry industry. Implementation of EU-Directive
on the “Welfare of Laying Hens” will increase egg production costs in the EU by 12% in barn
systems and even 20% in free range systems (GAIN Report, 2005), which was recently
confirmed by Bagnara (2009). Bagnara (2009) estimates that abolition of the traditional
battery cages by 1 January 2012 will reduce European productive potential by 40% as smallscale layer farmers will not be able to invest in aviary systems or enriched cage systems. If
similar standards are not introduced in non-European countries as well by legislation or
initiated by consumer demands, European legislations and regulations surely will reduce the
competitiveness of the European poultry industry and impact world trade on processed
poultry products. As welfare-based labelling might inform European consumers on the level
of animal welfare for animal products, an EU funded research program “Welfare Quality”
was started deliver the basics for such science-based welfare labelling.
II. EU-DIRECTIVES ON ANIMAL WELFARE
The EU-Directive on “Welfare of Laying Hens” (1999/74/EC) regulates the ban on traditional
battery cages by 1 January 2012. This Directive specifies a minimum floor space per hen
(being 750 cm2 in enriched cages and 1100 cm2 in aviary systems), availability of perches,
laying nests and an environment in which hens are able to show their natural behaviour. In
1
Schothorst Feed Research, PO Box 533, 8200 AM Lelystad, The Netherlands.
17
Aust. Poult. Sci. Symp. 2010...21
2007 68.6% of the eggs in the EU-27 were still produced by hens in traditional cages,
however, showing a large variation among EU countries (IEC, 2007). Also regulations with
respect to beak treatment of laying hens to prevent feather pecking and cannibalism is highly
variable among EU-27 countries as indicated by Van Horne and Achterbosch (2008). Recent
observations in the Netherlands indicated that floor-housed hens are more vulnerable to
enteric disorders like chronic enteritis, which was observed in 10% of floor-housed layer
flocks (De Bruijn et al, 2007). Chronic enteritis resembles focal duodenal necrosis, although
involvement of Clostridium colinum was not (yet) confirmed in the Netherlands. It was
estimated that such disorders might cost 2€/hen due to extremely high feed intakes and poor
laying persistency.
The EU-Directive on “Welfare of Meat Chickens” (2007/43/EC) details farming
conditions for broilers with respect to litter, feeders and drinkers, minimum length of a daily
dark period and maximises stocking density in broiler barns at 33 kg/m2. Higher stocking
densities are allowed up to 39 kg/m2 based on broiler management and housing and on
climate control (ambient temperature and air humidity, CO2 and NH3 levels at bird level). The
highest stocking density of 42 kg/m2 can only be used when broiler mortality does not exceed
a set maximum value for seven successive flocks being 1% plus 0,06% per day (which equals
3.4% during a 40 days growth period). Moreover, national guidelines have to be implemented
on good management practise (e.g. limiting the occurrence of hock burns and foot pad
dermatitis). This Directive will be effective by 1 July 2010. Although is introduced to
improve broiler welfare, it but will have (initial) adverse effects profit at farm level and be a
challenge for all parties of the broiler chain to improve broiler vitality and reduce mortality.
Apart from these EU-Directives, also additional national regulations are implemented e.g.
to prevent welfare problems related to chronic hunger stress in broiler breeders farming.
III. EU-DIRECTIVES ON POLLUTION PREVENTION
Several EU-Directives are implemented to control and reduce environmental pollution. EUDirective 2008/1/EC on Integrated Pollution Prevention and Control (IPPC), requires
industrial and agricultural activities to have a permit to produce, ensuring that environmental
conditions are met and companies use Best Available Techniques (BAT) to reduce and
prevent any pollution they may cause. IPPC applies to poultry farms with more than 40.000
bird places. Two other important environmental directives are the Nitrates Directive
(91/676/EEC) that maximises the application of N from animal manure to max. 170 kg N/ha
to control nitrate leakage from organic sources to ground and surface water and Directive
2001/81/EC on National Emission Ceilings (NEC) which sets upper limits for total emissions
in 2010 on SO2, NOx, volatile organic compounds and NH3 for each EU member state.
National regulations as part of the implementation of the NEC Directive limit e.g. the total
production of ammonia per bird place per year for broilers and laying hens. Nutritional
measures are part of the available techniques to reduce N and P excretions and ammonia
emission.
18
Aust. Poult. Sci. Symp. 2010...21
In Table 1 the effects of different nutritional measures are quantified to reduce the
excretion of N and P in animal manure.
Table 1.
Reducing concentrations of N, P and DM in animal manure through
feed supplements and feeding programs (adapted from Nahm, 2002,
2007 and Ferket, 2002).
Factor
Free amino acids and reduced protein
intake
per % CP reduction
Enzymes, general
Phytases
Growth promoters
Phase feeding
Choice of highly digestible feedstuffs
% reduction in manure
N
P
DM
18-35
10-27
8.5
5
12-15
5
25-35
5-30
5
10-33
10-25
5
5
Species
pullets and layers
broilers
Broilers
Broilers
Broilers
Broilers
Broilers
Broilers
IV. ALTERNATIVES FOR ANTI-MICROBIAL GROWTH PROMOTERS
Costs for intestinal disorders in broiler production can add up to 10€ per 100 broilers based on
a reduced body weight gain by 5% and increased feed conversion ratio by 3% (Table 2),
which estimates have been based on Schothorst Feed Research trials comparing production
performances of broilers with and without feed antibiotics. Costs for housing, litter, manure
and health care were obtained from statistics of the farmers’ average in 2009.
Table 2.
The costs of intestinal disorders in broilers (€ per 100 broilers), recalculated
from Van der Klis and Veldkamp (2007).
Amount
Sales profit
Body wt (kg)
Costs chicks and feed
Chicks
Feed (kg)
Feed profit (€/100 birds)
Variable costs
Housing, litter, manure
Health care
Gross profit (€/100 birds)
Healthy broilers
Price/unit
Total
Intestinal disorders
Amount
Total
220
0.83
182.60
209
173.47
100
352
0.28
0.32
-28.00
-112.64
41.96
100
344
-28.00
-110.08
35.39
-14.00
-2.00
25.96
-15.65
-4.00
15.74
It has been demonstrated in Schothorst Feed Research experiments that inducing
necrotic enteritis in broiler chickens via a successive Eimeria maxima/Clostridium
perfringens challenge reduces daily feed intake due to an acute phase response (Figure 1).
Comparing infected non-treated and non-infected non-treated indicated a decrease in albumin
(10.5 vs 11.7 g/L, P<0.05) and numerically increase in AGP (382.6 vs 318.9 mg/L) and
Ceruloplasmin (71.7 vs 54.4 mg/L) as shown by Ionescu et al. (2008). Niewold (2007)
hypothesised that alternatives for feed antibiotics should be selected based on their nonantibiotic anti-inflammatory response, rather than on their antibiotic effects. In hens affected
by chronic enteritis also clear positive effects were shown by anti inflammatory additives
(data not shown).
19
Aust. Poult. Sci. Symp. 2010...21
Figure 1.
The effect of a successive Eimeria maxima and Clostridium perfringens (Cp)
challenge on daily feed intake in broilers.
130
Daily feed intake (g/bird)
110
control
90
E. maxima
70
C. perfringens
E. max plus Cp
50
30
E. maxima
10
7
C. perfringens
14
21
age (days)
V. NUTRITION OF BROILER BREEDERS AND BROILER VITALITY
National regulations that do not allow skip-a-day feeding programs in broiler breeder nutrition
to prevent chronic hunger stress, being the case in the Netherlands, require different
approaches to limit nutrient intake and body weight gain of broiler breeders. Dietary dilution
of a normal density diets during rearing and laying (resp. 2600 and 2800 kcal/kg) by 10% and
20% increased eating time (De Jong et al., 2005), delayed maturation of the reproductive tract
by two weeks (Enting et al., 2007a), improved white to yolk ratio in eggs and growth of the
area vitellina externa and embryonic development, day-old chick weight of young breeder
hens and offspring body weight gain, whereas mortality was low (3.1%) and not affected by
dietary treatments (Enting et al. 2007c). Results are shown in Table 3.
Table 3.
Effect of dilution of broiler breeder diets on eating time, production
performance, maturation of the reproductive tract and offspring performance.
ND
ND-10%
ND-20%
Eating time of hens, min/day
6 wk
45c
75b
115a
c
b
20 wk
28
53
68a
c
b
30 wk
306
441
585a
Weight oviduct,g
24 wk
24,6
23,4
15,9
∆ wk 24-26
30,0b
37,6ab
53,3a
BW of 26-wk-old hen at, g
3106
3115
3106
Measurements on eggs of 29 wk old hens
White: yolk ratio, 48 hoi
1,626b
1,801a
2
b
Area vitellina externa (mm ), 48 hoi
700
910a
Embryo wt (g), 264 hoi
1,67b
2,04a
Offspring performance from eggs of 29 wk old hens
Wt one-day-old broiler, g
35,9
36,6
36,3
Wt 38-day-old broiler, g
2125b
2185a
2131b
ND: Normal Density; ND-10%: ND diluted by approx 10%; ND-20%: ND diluted by approx. 20%.
Hoi: hours of incubation; a,b,c Mean values with different superscript letters within a row differ
significantly.
20
Aust. Poult. Sci. Symp. 2010...21
Dietary dilution of breeder diets therefore seems not only to be an alternative for skip-a-day
feeding programs, it also improves embryonic development and offspring performance of
young broiler breeders.
REFERENCES
Bagnara GL (2009) Poultry Welfare Symposium, pp 154-161. Cervia, Italy.
DeBruijn N, van der Klis JD, Lensing M, Fabri T (2008) Chronic enteritis: A problem in
Floor-housed laying hens. Proceedings, World Poultry Congress. Brisbane, Australia.
DeJong IC, Enting H, van Voorst A, Blokhuis HJ (2005) Poultry Science 84, 194-203.
Enting HA, Veldman A, Verstegen MWA van der Aar PJ (2007a) Poultry Science 86, 720726.
Enting HA, Kruip TAM, Verstegen MWA, van der Aar PJ (2007b) Poultry Science 86, 85856.
Enting HA, Boersma WJA, Cornelissen JBWJ, van Winden SCL, Verstegen MWA, van der
Aar PJ (2007c) Poultry Science 86, 282-290.
Ferket PR, van Heugten E, van Kempen TATG, Angel R (2002) Journal of Animal Science
80, E168-E182.
GAIN Report (2005) Abolition of battery cages to cost €354 million to EU-25 egg producers.
USDA Foreign Agricultural Service. GAIN Report E35065.
IEC (2007). Comparison of international country data. International Egg Market. Annual
Review. International Egg Commission. London.
Ionescu C, Bravo D, Lensing M, van der Klis JD, Niewold T, Bruggeman V (2008) Poultry
Science 87, 74 (Abstr).
Nahm KH (2002) Critical Reviews in Environmental Science and Technology 32, 1-16.
Nahm KH (2007) World’s Poultry Science Journal 63, 625-654.
Niewold TA (2007) Poultry Science 86, 605–609.
Van der Klis JD, Veldkamp T (2007) International Poultry Production 5, 11-12.
Van Horne PLM, Achterbosch TJ (2008) World's Poultry Science Journal 64, 40-52.
21
Aust. Poult. Sci. Symp. 2010...21
UPDATE ON CURRENT EUROPEAN BROILER BONE PROBLEMS
C. C. WHITEHEAD 1
Summary
Changes in genetic and management practices seem to be resulting in new patterns of bone
problems in broilers on European farms over the last 2 years. Problems associated with
endochondral bone growth include femoral head necrosis (FHN), rickets and tibial
dyschondroplasia (TD). FHN seems to be the most prevalent and widespread problem
involving bacterial infection that can be exacerbated by abnormalities of epiphyseal cartilage.
Attention to hatchery hygiene seems to be important in trying to counter this problem.
Rickets resulting from dietary errors with calcium, phosphorus or vitamin D or malabsorption
of these nutrients is an occasional problem. TD still occurs, despite attempted selection
against it. Dietary supplementation with 25-hydroxyvitamin D is the most effective dietary
means of prevention. Black bone syndrome (BBS) is a recently identified problem that results
from leakage of blood through porous areas of bone, particularly near the proximal tibia. The
blood can darken during processing, particularly freezing of leg portions and can even spread
into surrounding meat, causing consumer acceptance problems. Use of dietary 25hydroxyvitamin D appears to be a promising approach to improving bone structure and
minimising blood leakage in BBS.
I. INTRODUCTION
Modern intensively-reared broiler strains have been selected over many generations primarily
for fast growth of muscle tissues combined with good feed efficiency. Concentration of
selection effort on these two characteristics resulted in increasing incidences of defects in
other physiological areas. Leg defects in particular became quite prevalent and led breeders to
pay more attention to skeletal characteristics in their selection practices in recent years. The
result has been, with the odd notable exception, a general improvement in skeletal
characteristics of the more recent broiler strains. Changes in management practices, aimed at
improving leg health, have also been proving effective. The net result of these improvements
in genetic and management procedures has been a general enhancement in bone quality and
leg health and welfare of broilers. Based on my recent observations in commercial broiler
flocks around Europe, the greatest improvement seems to be in the non-specific rotational
and angular deformities, of the valgus/varus type. However, bone problems have not yet been
abolished. The increasingly greater growth rates and feed efficiencies of broilers put ever
greater emphasis on optimising both nutritional and hygienic practices. Moreover, there have
been genetic changes in bone structures, and these may need to be met by changes in nutrient
specifications. There is also increasing evidence that some of these structural genetic changes
may be resulting in new bone problems. The net result is that bone problems still occur,
though their pattern is changing, and continued genetic, nutritional and management
vigilance is needed to keep them under control.
1
Roslin Institute, Roslin, Midlothian EH25 9PS, UK ([email protected])
22
Aust. Poult. Sci. Symp. 2010...21
II. LEG PROBLEMS ASSOCIATED WITH ENDOCHONDRAL BONE GROWTH
a) Femoral head necrosis (FHN)
This seems to be currently one of the most widespread broiler leg problems. Strictly
speaking, it is not a nutritional problem since it is a bacterial osteomyelitis associated mainly
with Stappylococcus aureus, though sometimes with Eschericia coli. These bacteria can
become trapped in the growth plate blood vessels, especially in the femoral head where blood
flow may be weakest. But they may also become trapped in damaged or abnormal cartilage
associated with nutritional problems such as rickets or tibial dyschondroplasia (TD). The
infection spreads initially to the chondrocytes, causing necrosis of these cells, before
extending into the bone itself. The condition is painful and affected birds can be identified by
their extreme reluctance to stand or walk. FHN can be confirmed by attempted dislocation of
the hip joint. In mild cases, epiphyseolyis takes place, in which the bone of the proximal
femur separates from the cartilage cap, leaving the latter attached to the joint socket. In more
severe cases, the bone of the proximal femur fractures.
An important solution to the problem seems to involve better hatchery hygiene, since
a strong association has been shown between hatchery of origin and occurrence of FHN. The
appropriate handling of dirty eggs is probably a key factor in minimising the likelihood of
spreading infection among hatching chicks but avoidance of cartilage defects linked to
nutrition in the growing broilers is also helpful. Antibiotic treatment is often tried, but seems
to have only limited impact in preventing the condition.
A possible reason for the widespread occurrence of FHN could be a continuing
reduction in the immune responsiveness of broilers as a consequence of further selection for
faster growth. Perhaps the use of nutrients such as vitamins D and E to enhance immune
status is an approach that could be considered to counter this tendency. Avoidance of
immunosuppressive diseases such as infectious bursal disease would also be important.
b) Rickets
Deficiencies of calcium, phosphorus and/or vitamin D resulting in rickets in young birds still
occur occasionally. The morphology of the growth plate can be used to distinguish between
the rickets caused by calcium/vitamin D deficiency or phosphorus deficiency. In the former,
there is a considerable increase in the thickness of the proliferative zone. In phosphorus
deficiency rickets, the proliferative zone is relatively normal in size but there is a
considerable increase in the size of the hypertrophic zone and a delay in mineralisation of this
zone. Rickets of both types can be caused by dietary deficiency, an imbalance in the
proportions of calcium and phosphorus or the use of feed additives that may inadvertently
impair nutrient absorption. Rickets can also occur with diets that are not overtly inadequate in
calcium/phosphorus/vitamin D and in these cases the problem is thought to be caused by
infections that impair nutrient absorption (malabsorption).
c) Tibial dyschondroplasia (TD)
TD is primarily a genetic problem that can be influenced by nutrition. Breeders can select
against TD, using a Lixiscope, an x-ray fluorescence device. This device can detect large TD
lesions, though small lesions are less easy, and its use has resulted in a decrease in the general
incidence of TD in some broiler strains. However, TD still occurs in commercial broiler
production.
Several nutritional factors can influence the occurrence of TD. A low ratio of Ca:P
can increase the incidence of TD, but there is no indication that a correct balance of these
nutrients will necessarily prevent the condition (Edwards and Veltmann, 1983). Other
nutrients, including monovalent cations and anions (Na+, K+, Cl-) can also alter the
23
Aust. Poult. Sci. Symp. 2010...21
incidence, but the only nutrients that have so far been shown to completely prevent TD are
vitamin D and its metabolites. 1,25-dihydroxyvitamin D (1,25-D) is particularly effective in
the dose range 5 to 10 mg/kg but has potential practical hazards from toxicity (Rennie et al.,
1995). 25-hydroxyvitamin D has also shown to be effective, though at higher dose rates (75250 mg/kg) (Rennie and Whitehead, 1996). It has a much higher toxic threshold than 1,25-D
and is available commercially as HyD.
More recent findings (Whitehead et al., 2004) indicate that vitamin D itself, when fed
at high dose rates, can reduce or prevent the occurrence of TD at 14 d of age. Basal diets of
different Ca and available P (avP) contents were supplemented with different concentrations
of vitamin D, giving different incidences of TD. TD incidences responded to vitamin D
supplementation, but concentrations above 5000 IU/kg were needed to minimise TD
incidence. These results suggest that the vitamin D3 requirements of young broilers can be
quite high and that responses in performance and bone characteristics can occur to dietary
concentrations above 5000 IU/kg at both conventional and abnormal dietary calcium and
phosphorus concentrations or ratios up to 14 days. Practical vitamin D3 supplements are
normally in the range 3000-5000 IU/kg but, in view of the current findings, it would seem
prudent to supplement at the higher end of this range or above if regulations permit,
especially in broiler starter diets. Higher dietary vitamin D potencies can be obtained by
replacement of a proportion of the feed vitamin D3 supplement with 25-hydroxyvitamin D3
(e.g. 3000 IU vitamin D3 and 50 µg 25-hydroxyvitamin D3/kg). Alternatively, additional
vitamin D3 or 25-hydroxyvitamin D3 can be administered via the drinking water.
III. PROBLEMS OF INTRAMEMBRANOUS OSSIFICATION
The black bone syndrome
The blackening of broiler leg bones and the spreading of the discolouration into adjacent
meat, especially after cooking, has been reported intermittently in recent years (e.g. Lyon and
Lyon, 2002; Saunders-Blades and Korver, 2006). However, more recent observations on
broiler processing lines and supermarket products are suggesting that the problem is much
more widespread and fundamental than has been appreciated. It has implications for both
broiler meat product quality and bird welfare and is now being described under the name
black bone syndrome (BBS).
BBS seems to be associated with leaking of blood from the marrow through the bone
of the proximal tibia and can be observed in freshly killed birds on processing lines, most
frequently in the tibia. The problem of BBS seems to be made worse by freezing un-deboned
leg portions. Presumably the freezing process forces more blood from the marrow through the
bone and into adjoining meat. Cooking may contribute further to this process and also
blackens the blood. The final result is a blackened and unappetising appearance of the meat
surrounding the bone. This is already having market impacts because there are now reports
that fast-food outlets are stopping using frozen broiler leg portions or using deboned meat
instead.
BBS seems to be caused by a problem in intramembranous ossification. Cortical bone
structure is becoming more porous in modern broilers (Williams et al., 2000). This structure
in the midshaft is still strong and able to bear the weight of the bird. However, we have found
a different situation in the shaft of the bone immediately adjacent to the proximal growth
plate, the area through which the blood is seen to be leaking. Here, the bone is much thinner –
it can often be depressed by moderate finger pressure - and we have confirmed this by
histological examination. The overall mineralised area is very thin and comprised of poorly
connected strands of bone, of more cancellous than cortical appearance. There are obvious
routes in this structure through which blood could seep and collect under the periosteal layer.
24
Aust. Poult. Sci. Symp. 2010...21
The long-term solution to the problem would appear to require breeding companies to
pay more attention to bone structure in their selection of future breeding stock. However, in
the short-term it is possible that nutritionists might be able to help alleviate the problem by
paying attention to factors that can maximise bone quality. We have followed this up by
comparing various characteristics of bone from another study on the feeding of HyD carried
out by IRTA in Spain. The treatments involved supplementation of the diets of broilers with
either 0 or 69 µg 25-hydroxyvitamin D (as HyD), in addition to a normal dietary content of
vitamin D3. We measured the reflected optical density (as a measure of bone blackening) in
the area of 10 to 15 mm from the proximal ends of frozen tibias. This measure of blackening
was significantly less (P < 0.05) in birds given the HyD supplement. It is possible that further
research may identify improved calcium and phosphorus nutritional practices which, when
combined with the use of HyD, may further contribute to the alleviation of the BBS problem.
REFERENCES
Edwards HM, Veltmann JR (1983) Journal of Nutrition 113, 1568-1575.
Lyon BG, Lyon CE (2002) Poultry Science 81, 1916-1920.
Rennie JS, Whitehead CC (1996) British Poultry Science 37, 413-421.
Rennie JS, McCormack HA, Farquharson C, Berry JL, Mawer EB,Whitehead CC (1995)
British Poultry Science 36, 465-477.
Saunders-Blades J, Korver D (2006) European Poultry Conference Verona, Italy.
Whitehead CC, McCormack HA, McTeir L, Fleming RH (2004) British Poultry Science 45,
425-436.
Williams B, Solomon S, Waddington D, Thorp B, Farquharson C (2000) British Poultry
Science 41, 141-149.
25
Aust. Poult. Sci. Symp. 2010...21
UTILISATION OF METABOLISABLE ENERGY OF FEEDS IN PIGS AND POULTRY:
INTEREST OF NET ENERGY SYSTEMS?
J. NOBLET 1, J. VAN MILGEN1 and S. DUBOIS1
Summary
The evaluation of the energy content of pig or poultry feeds has been most commonly based
on their DE or ME contents. However, the closest estimate of the "true" energy value of a
feed should be its NE content, which takes into account differences in metabolic utilization of
ME of nutrients. This review first considers some methodological aspects of NE
determination. Experimental data in pigs indicate that the NE/ME ratio varies greatly with the
chemical composition of diets and nutrient, with ratios for fat (90%) and starch (80%) that are
higher than for protein and dietary fibre (60%). This has marked consequences on the relative
energy values of ingredients according to the energy system that is used. Consequently, the
NE system is better in predicting the performance of pigs. With regard to poultry, the ranking
between nutrients for NE/ME is similar to what is observed in pigs but with smaller
differences between nutrients. This is consistent with results of energy balance trials that are
unable to detect significant differences in NE/ME between diets that differ markedly for their
chemical composition. In any case, the accuracy of the NE value is highly dependent on the
accuracy of DE or ME values or digestible nutrient contents. Overall, there is an obvious
advantage in using NE systems for pigs while further investigations are required for
implementing a reliable NE system for poultry.
I. INTRODUCTION
The cost of feed is the most important cost of poultry or pig meat production (~60%) and the
energy component represents the greatest proportion of this cost. Therefore, it is important to
estimate precisely the energy value of feeds, either for least-cost formulation or for adapting
feed supply to the energy requirements of animals. Evaluation of the energy content of pig or
poultry feeds has been most commonly based on their DE or ME contents. However, the
closest estimate of the “true” energy value of a feed should be its NE content, which takes
into account differences in metabolic utilization of ME of nutrients for maintenance and
production requirements. In addition, NE is the only system in which energy requirements
and diet energy values are expressed on a same basis which should theoretically be
independent of the feed characteristics. In many parts of the world, NE systems have been
implemented in practical pig nutrition. On the other hand, NE systems are very little used for
poultry. The objective of this review paper is to consider the recent contributions regarding
the efficiency with which ME is used in both pigs and poultry and to suggest proposals for
future evaluation of energy in poultry nutrition. More complete information can be obtained
in recent reviews (Pirgozliev and Rose, 1999, Noblet and van Milgen, 2004; Noblet, 2006).
The effects of feed characteristics (physico-chemical composition), animal factors such as
body weight, physiological stage or species (in poultry productions) or technological factors
such as particle size, pelleting, extrusion or the addition of enzymes, that affect primarily
digestion will not be considered here (see Le Goff and Noblet, 2001; Noblet and Le Goff,
2001; Noblet and van Milgen, 2004; Noblet, 2006 for more information).
1
INRA, UMR1079 SENAH, F-35590 – St-Gilles, France
26
Aust. Poult. Sci. Symp. 2010...21
II. METHODOLOGICAL ASPECTS
Not all gross energy (GE) of a feed is available for meeting the requirements of animals since
variable proportions of GE are lost in faeces, in urine, as fermentation gases (i.e., methane,
hydrogen) and as heat (or heat increment, HI). The DE content of a feed corresponds to its
GE content minus faecal energy losses after digestion in the digestive tract. Even though they
are related to digestion, energy of gas and heat originating from hindgut fermentation is not
considered in the calculation of DE. The ME content of a feed corresponds to the difference
between the DE content and energy losses in urine and gases. Most of the energy lost in gases
is due to methane production, which typically is very small in growing pigs and poultry, but
not so in adult sows (Le Goff et al., 2002). Consequently, most ME values in literature and
tables for growing pigs and poultry ignore energy losses as methane.
Net energy is defined as the ME content minus HI associated with feed utilization (i.e.,
energy cost of ingestion and digestion and HI related to metabolic utilization of ME) and the
energy cost corresponding to a "normal" level of physical activity (figure 1). However, the HI
of a given feed (as % of ME) may not be constant over a large range of ME intakes for a
given animal and depends on several physiological factors. For instance, the HI tends to be
lower below than above maintenance energy supply (Noblet et al., 1993; 1994; 1994). The HI
is also lower when ME is used for fat deposition compared with protein deposition (Noblet et
al., 1999). Therefore, for comparing different feeds for their HI or the efficiencies of ME
utilization, it is necessary to calculate these values under similar conditions (e.g., feeding
levels), at protein and amino acid supplies meeting the requirement and/or constant
composition of the gain and/or at a given physiological stage. This was implemented in the
studies of Fraps (1946) in broilers or Noblet et al. (1994) in growing pigs.
Measurement of energy balance and heat production (HP) is rather complex and timeconsuming so that animals are usually fed at only one "operational/practical" feeding level.
For a growing animal, the NE intake is then calculated as the sum of retained energy (RE) at
this feeding level and fasting heat production at zero activity (FHP) (Noblet et al., 1994). This
NE value or the efficiency of ME for NE (k as NE/ME) then corresponds to a combined
utilization of energy for meeting requirements for maintenance and for growth. The value of
FHP is either measured directly in fasted animals or obtained from literature. It can also be
calculated by extrapolating HP measured at different feeding levels to zero ME intake.
However, this latter method, even though it has been widely used in the past, has important
limitations, mainly because the growing animal adapts its FHP to its feeding/energy level
prior to fasting (Koong et al., 1982; de Lange et al., 2006; Labussière et al., 2008) and
questions the use of extrapolated HP at zero ME intake for getting an estimate of FHP.
Consequently, direct measurements of FHP are preferable; the use of literature measurements
of FHP is an alternative solution. In any case, the NE value is dependent on the FHP value.
From a practical point of view and to avoid bias in the calculation of NE for a series of
feeds, it is necessary to carry out energy balance measurements in similar animals (i.e., same
sex, same breed and the same body-weight range), to keep these animals in a temperaturecontrolled environment within their thermoneutral zone, to minimize variation in behaviour,
and to feed the animals at the same energy level with balanced diets so that the animals can
express their production/growth potential. Under these circumstances, an erroneous estimate
of FHP will affect the absolute NE value but not the ranking between feeds. This also means
that NE should not be measured in animals fed ingredients for which the chemical
characteristics are very different from those of a complete balanced diet.
Heat production can be measured directly through direct calorimetry, estimated from gas
exchanges through indirect calorimetry or calculated as the difference between ME intake
and energy gain, this latter component being measured according to the comparative
27
Aust. Poult. Sci. Symp. 2010...21
slaughter technique (CST). The most commonly used method is indirect calorimetry, which
consists in calculating HP mainly from oxygen consumption and carbon dioxide production,
and to a lesser extent from metabolic nitrogen and gas energy losses (Brouwer, 1965). This
method also allows measurements over a short period of time (i.e. a few days) with
possibilities of combination of measurements at different energy levels (including fasting) on
the same animal without adaptation. In addition, modelling methods can be implemented to
partition the total HP between different components, which can be used in the further
interpretation of energy balance data (van Milgen et al., 1997; figure 1).
9
8
Heat
production,
kJ/min
Activity HP
TEF short
TEF long
Fasting HP
7
Heat
increment
6
5
4
3
2
1
0
8:25
10:25
12:25
14:25
16:25
18:25
20:25
22:25
0:25
2:25
4:26
6:26
Time
Figure 1.
Dynamics of components of heat production in a group of 9 broilers (1 kg
BW) fed six meals per day (INRA data; HP: heat production; TEF: thermic
effect of feed)
While measurements of DE and, to a smaller extent, of ME are easy and can be
undertaken on a large number of feeds at a reasonable cost, the measurement of NE is far
more complex and expensive. The best alternative is then to use information from prediction
equations that were obtained from experiments carried out under similar and standardized
conditions. In our laboratory, we measured NE on a wide range of different diets and
proposed NE prediction equations, allowing to estimate the NE value of ingredients and
complete diets based on available information of DE or ME content, combined with
information of chemical characteristics (Noblet et al., 1994a). The latter information can be
obtained from existing feeding tables or digestibility trials. The prediction equations obtained
from these measurements can then be applied to any type of feed without further NE
measurements (Noblet and van Milgen, 2004; Noblet et al., 2004).
III. UTILIZATION OF ME IN PIGS
Over the last 50 years, several experiments have been carried out by different laboratories to
quantify the effect of diet and animal factors on HI or k in pigs (see the review of Noblet,
2006). The most recent and complete study was carried out by INRA with measurements on
61 diets (Noblet et al., 1994; Noblet, 2006). From their trials and other results, these authors
28
Aust. Poult. Sci. Symp. 2010...21
showed that the maintenance energy requirement (and FHP) in growing pigs is proportional
to BW0.60, and not to the commonly used metabolic BW (BW0.75) (Noblet et al., 1999)). The
FHP at thermoneutrality and zero activity averaged 750 kJ/kg BW0.60/d. On this basis, the
efficiency of ME for NE in growing pigs (kg, %) averaged 74% but varied with chemical
characteristics (g/kg DM) according to the following equation:
kg = 74.7 + 0.036 x EE + 0.009 x Starch – 0.023 x CP – 0.026 x ADF (RSD = 1.2) 2.
A similar equation was proposed for adult sows fed at their maintenance energy level (Noblet
et al., 1993). The variation in kg is due to differences in efficiencies of ME utilization
between nutrients with the highest values for fat (~90%) and starch (~82%) and the lowest
(~60%) for dietary fibre (DF) and CP (Noblet et al., 1994). These values were confirmed
experimentally in our laboratory (van Milgen et al., 2001). Measurements conducted in pigs
having different BW and composition of BW gain, suggested that the efficiency of ME for
NE was little affected by the composition of BW gain, at least under most practical
conditions (Noblet et al., 1994). Similarly, the ranking between nutrients for their efficiencies
was similar in adult sows fed at maintenance level and in growing pigs. Finally, the heat
increment associated with protein utilization, either retained as protein or as lipid, appeared to
be constant (van Milgen et al., 2001), which means that the NE value of dietary CP does not
depend on its final utilization.
IV. UTILIZATION OF ME IN POULTRY
As for pigs, several trials or theoretical assumptions over the last 70 years have carried out to
quantify the utilization of ME and its digestible nutrients for NE in poultry (see the review of
Pirgozliev and Rose, 1999). Some studies were carried out on ingredients or unbalanced diets
with some subsequent limitations in the interpretation of the results. The most comprehensive
series of measurements were conducted in USA by Fraps (1946) and the Rostock group
(Schiemann et al., 1972), mainly focussing on feed ingredients. Unfortunately, most attention
in these studies was focused on starch, DF and CP with little variability in fat content of diets
or feedstuffs. A recent study of Carré et al. (2002) was conducted on complete feeds (n=28)
fed to 3 to 5 wk old broilers while varying the nutrient composition of the diets. The CST was
used to quantify energy retention and NE was calculated according to a FHP value (500 kJ/
kg BW0.60/d) measured in respiration chambers (van Milgen et al., 2001). Later studies carried
out at INRA suggested that FHP in 0.5 to 3.0 kg broilers is proportional to BW0.70 with values
ranging between 420 and 450 kJ/kg BW0.70/d (Warpechowski et al., unpublished data).
In several literature studies, efficiencies of DE or ME for NE for maintenance + growth
(or fattening) have been quantified, and some of these values are listed in table 1. The
NE/ME ratio can also be calculated for a large series of feedstuffs in the studies of Fraps
(1946). As for pigs, the lowest efficiency is observed for CP and the highest for fat. However,
the difference between the most extreme values (i.e., CP and EE) appears to be somewhat
lower in poultry (65 to 85%) than in pigs (60 to 90%). In addition, the values from
Schiemann et al. (1972) were obtained in adult fattening birds, which are in a physiologically
different stage while genotypes differed from modern birds. From that point of view, the
study of Carré et al. (2002) is probably more representative of modern broiler production.
2
EE: ether extract, CP: crude protein; ADF: Acid Detergent Fiber
29
Aust. Poult. Sci. Symp. 2010...21
Table 1.
Efficiencies of ME from digestible nutrients for NE in poultry (%)
Reference
Schiemann et al., 1972
De Groote, 1974
Carré et al., 2002
1
Production1
M + fattening
M + growth
M + growth
CP
61
60
68
EE
84
90
84
Carbohydrates
75
75
78
Diet
732
742
76
M: maintenance 2Assuming that 25, 20 and 55% of ME is provided as CP, EE and carbohydrates
An alternative approach to study the effect of diet on the efficiency of ME for NE was
tested in recent trials conducted at INRA (Noblet et al., 2007; 2009). The approach consisted
in preparing diets focussing each time on one specific nutrient (in exchange for starch). The
effects of CP, EE and DF contents were evaluated and the trials were conducted in respiration
chambers in group housed and ad libitum fed growing broilers over the 3rd to 7th week of age.
The methods developed by INRA were implemented to partition total HP between its main
components (figure 1). The most important results are presented in Table 2.
Table 2.
Trial
1
2
3
4
Effect of diet composition on heat production and efficiency of ME for NE in
broilers: compilation of INRA data1
Diets
18% CP
22.7% CP
22.5% CP
27.3% CP
2.8% EE
9.7% EE
9.5% NDF
17.7% NDF
kJ/kg BW0.70/d
HP
853
846
892
872
904
901
912
923
ME
1609
1609
1457
1457
1873
1877
1503
1521
NE/ME
%
75.1
74.8
67.7
68.6
75.0
75.7
71.3
72.3
AHP
146
153
173
168
141
152
170
175
1
Measurements carried out in groups of broilers weighing 1.3 to 1.5 kg BW on average for each trial; the
indirect calorimetry method in respiration chambers was used; AHP: Activity heat production; complementary
details by Noblet et al. (2007) for trials 1 and 2 and Noblet et al. (2009) for trial 3; trial 4: unpublished data. In
trials 1, 2 and 3, the variation in CP or EE is associated with an inverse variation in starch content; in trial 4, the
increased NDF level corresponds to a dilution by dietary fibre provided by wheat bran, maize bran and soybean
hulls. In trials 1 and 2, data have been adjusted for a similar ME intake while observed values are given for trials
3 and 4. None of the differences between treatments within each trial were significant (P>0.05).
Table 3.
Comparative effect of dietary crude protein on energy utilisation in growing
pigs and broilers (adapted from Noblet et al., 2003)1.
Species
Dietary protein level
Body weight, kg
Energy balance, kJ/kg BW0.60
ME intake
Heat production
Fasting heat production
Heat increment2
NE/ME (x100)
Pigs
Normal
57.6
Low
57.2
2564
1402a
735
667a
73.9a
2566
1346b
731
614b
75.9b
1
Broilers
Normal
Low
1.47
1.46
1626
862
446
417
74.8
1642
861
456
404
75.0
The reduction of dietary CP consists in a replacement of protein from soybean protein concentrate by maize
starch with supplementation of amino acids in order to meet the requirements. 2sum of TEF and AHP (see
Figure 1)
30
Aust. Poult. Sci. Symp. 2010...21
The most surprising result of these trials is the absence of effect (within a trial) of dietary
CP on HP, HI and NE/ME in broilers; a parallel study conducted in both pigs and broilers
confirmed this major difference between pigs and poultry (Noblet et al., 2003; Table 3).
Table 2 also indicates that the replacement of starch by fat was unable to increase
significantly the NE/ME ratio of the diet (Table 1). Either a higher difference in fat content
between diets and/or a higher number of observations would be required for getting a
significant difference. Finally, the presence of high levels of undigested DF in broilers diets
did not significantly change the HP and the NE/ME ratio of the diet; this latter criterion was
even numerically higher with the high DF diet. Anyway, this result does not confirm the
hypotheses of Emmans (1994), who suggested attributing an energy cost to undigested
organic matter for its excretion in order to calculate the so-called "effective" energy.
Similarly, a recent study conducted in pigs did not indicate any additional energy cost of the
presence of undigested DF provided by wheat straw (de Lange et al., 2006). Overall and
unlike pigs, these studies indicate that major changes in diet composition do not affect the
efficiency of ME for NE in broilers. The possible effects of dietary fat may deserve further
studies, especially for an extrapolation of results to high fat single ingredients. This result
also means that the hierarchy between diets for poultry would not be markedly affected by
the energy system (ME vs. NE) and the design of trials such as those presented in table 2 did
not indicate significant differences in the efficiency of ME utilization due to nutrients.
V. NET ENERGY SYSTEMS
An energy system corresponds to a method of prediction of the energy value of compound
feeds and ingredients for a given type of animals. It then combines a step in energy utilization
(DE vs. ME vs. NE) and a calculation method. With regard to NE systems, most of them
combine the utilization of ME for maintenance and for growth or for fattening and they are
based on prediction equations taking into account either digestible nutrients or DE (or ME)
and some chemical characteristics (see reviews of Pirgozliev and Rose, 1999 for poultry and
Noblet, 2006 for pigs). Some systems have been established from measurements on animals
(Schiemann et al., 1972; Noblet et al., 1994; Carré et al., 2002) while others have been
proposed from literature data and/or biochemical information and/or theoretical assumptions
(Emmans, 1994). It should also be noticed that an adequate energy system is important to
consider since most other nutritional criteria (protein, amino acids, minerals, etc.), either for
feed value or for animal requirements are expressed relative to the energy content of the feed.
Table 4.
Relative DE, ME and NE values of ingredients for growing pigsa
Animal fat
Corn
Wheat
Reference diet
Pea
Soybean (full-fat)
Wheat bran
Soybean meal
DE
243
103
101
100
101
116
68
107
ME
252
105
102
100
100
113
67
102
a
NE
300
112
106
100
98
108
63
82
NE/ME, %
90
80
78
75
73
72
71
60
From Sauvant et al. (2004). Within each system, values are expressed as percentages of the energy value of a
diet containing 68% wheat, 16% soybean meal, 2.5% fat, 5% wheat bran, 5% peas and 4% minerals and
vitamins.
31
Aust. Poult. Sci. Symp. 2010...21
The system proposed by Noblet et al. (1994a) for pigs and applied in the INRA & AFZ
feeding tables (Sauvant et al., 2004) is based on a large set of measurements (61 diets). The
NE prediction equations that have been generated from these measurements include either
DE values and chemical characteristics or digestible nutrients as predictors; they are
applicable to ingredients and compound feeds and at any stage of pig production (Noblet,
2006). In this NE system and assuming that NE represents the best estimate of the "true"
energy value of feeds, the energy value of protein-rich or fibrous feeds is overestimated when
expressed on a DE (or ME) basis. On the other hand, fat or starch sources are underestimated
in DE and ME systems (Table 4). Unfortunately, there is no comparable information for
poultry, at least according to a general agreement on the prediction equations or systems to be
implemented. According to the NE/ME ratios provided by Carré et al. (2002) for the main
nutrients, the impact of moving from ME to NE for estimating the energy value of ingredients
should be similar to what is observed in pigs (table 4) but to a smaller extent. However, our
experimental results using indirect calorimetry failed to confirm this; additional
measurements may be needed.
It is important to point out that specific and accurate DE (or ME) values or digestible
nutrient contents should be used in the calculation of NE contents. For instance, energy
digestibility differs between growing pigs and adult sows, with subsequent different NE
values
of
feeds
for
both
stages
(Le
Goff
and
Noblet,
2001;
website: http://www.inapg.inra.fr/dsa/afz/tables/). In fact, reliable information on digestibility
of energy or of nutrients is the most limiting factor for predicting energy values of feeds for
pigs or poultry. The lack of comprehensive information on the effects of technology (e.g.,
pelletting, extrusion, enzymes addition) or about the differences in digestion between poultry
species or physiological stages (growing vs. adult, etc.) is a major limiting factor for getting
accurate estimates of energy values for pigs or poultry, irrespective of the energy system used.
VI. VALIDATION OF NET ENERGY SYSTEMS
An energy system, as any evaluation system for feeds, should be validated before being
implemented. With regard to a NE system, a first level of validation is to measure the NE of
feeds not included in the construction of the system and to compare these measured values to
those calculated from the prediction equations proposed by the system. An illustration of this
validation approach is given for the NE system proposed by Noblet et al. (1994a) for pig
feeds (Figure 2). This figure indicates that this NE system appears quite robust. A satisfactory
energy system should also be able to predict the energy value of any feed, especially
ingredients for which the chemical composition is largely different from the composition of
conventional ingredients or diets. The study by Noblet et al. (1993) on pigs confirmed that
the NE system established on compound feeds is applicable to individual ingredients. In a
review article concerning poultry, Pirgozliev and Rose (1999) compared the NE value of a
series of feedstuffs as measured by Fraps (1946) to the values calculated according to
different NE prediction equations. None of the literature NE systems for poultry was able to
accurately predict the NE values measured on the feedstuffs; the biggest discrepancy was
observed with the system proposed by Emmans (1994).
A last stage of validation of an energy system is its ability to predict the performance of
animals. From that point of view, NE systems which take into account the final stage of
energy utilization should be better able to predict animal performance than DE or ME
systems. In addition, it is important to use the same energy system for expressing the diet
energy values and the animal energy requirements; the energy system in which requirements
32
Aust. Poult. Sci. Symp. 2010...21
are the least dependent on diet characteristics is the NE system. In the case of pigs, the
superiority of the NE system as proposed by Noblet et al. (1994) has been clearly
demonstrated from results of growth trials. They indicate that the energy cost of growth or the
daily energy requirement are independent of diet composition when expressed on a NE basis.
On the other hand, on DE or ME bases, the energy cost is decreased when CP content is
decreased or fat content is increased (Table 5). In other terms, unlike the NE system, the
ability of DE and ME systems to predict the performance of pigs is lower than that of a NE
system. For poultry, this exercise is difficult since no NE system is widely accepted and
applied, so that there is no response trial in which BW or energy gain are related to either ME
or NE intakes. Recent data by van der Klis (personal communication) did not show any
superiority of a NE system partly based on the studies of Schiemann et al. (1972) on the ME
system. This conclusion would be in agreement with our balance studies indicating that the
NE/ME ratio was not significantly affected by diet composition (Table 2).
15
14
NE measured
y = 0.999x
13
12
11
10
9
9
10
11
12
13
14
15
NE calculated
Figure 2.
Relationship between measured NE values of compound feeds (n=41; indirect
calorimetry method) and NE calculated according to NE prediction equations
of Noblet et al., (1994) (adapted from INRA data)
Table 5.
Performance of growing-finishing pigs according to energy system and diet
characteristics1
Energy system
Trial 1: Added fat (%)
0
2
4
6
Trial 2: crude protein content (30-100 kg)
Normal
Low
Trial 3: crude protein content (90-120 kg)
Normal
Low
DE
ME
NE
100
100
99
98
100
100
99
98
100
100
100
100
100
96
100
97
100
100
100
97
100
98
100
100
1
Energy requirements (or energy cost of BW gain) for similar daily BW gain and composition of BW gain;
values are expressed relative to the energy requirement (or energy cost of BW gain) in the control treatment
(considered as 100; values in bold characters); from Noblet (2006) and unpublished data.
33
Aust. Poult. Sci. Symp. 2010...21
VII. CONCLUSION
This review indicates that NE is a better predictor than DE or ME of the "true" energy value
of poultry or pig feeds. Available information for pigs indicates an obvious interest of
formulating on a NE basis and NE systems should be implemented for getting a reliable
prediction of performance of animals. In the case of poultry, the conclusions are less clear
and convincing with no marked advantage of a NE system over a ME system for predicting
the performance of broilers; further investigations are necessary to evaluate the potential
interest of a NE system for poultry. Probably most attention should be focused on the impact
of fat level with possible consequences on the relative energy value of fat-rich ingredients
and subsequent results in least-cost formulation. Finally, even though NE is the final
objective in energy evaluation of feeds, most attention should be paid to the accurate
estimation of DE or ME values, which are the most important factors of variation of the
energy value of poultry or pig feeds.
REFERENCES
Brouwer E (1965) Proceedings, Energy Metabolism, [Blaxter KL Ed.] EAAP Publication N°
11, Academic Press, London (UK), 441-443.
Carré B, Lessire M, Juin H (2002) Proceedings, Eastern Nutrition Conference, Guelph
(Canada), 140-149.
De Groote G (1974) British Poultry Science 15, 75-95.
De Lange C, van Milgen J, Noblet J, Dubois S, Birkett S (2006) British Journal of Nutrition
95, 1082-1087.
De Lange C, van Milgen J, Dubois S, Noblet J (2006) Animal Research 55, 551-562.
Emmans G (1994) British Journal of Nutrition 71, 801-821
Fraps GS (1946) Texas Agricultural Experiment Station Bulletin n° 678.
Koong LJ, Nienaber JA, Pekas JC, Yen JT (1982) Journal of Nutrition 112, 1638-1642
Labussière E, Dubois S, van Milgen J, Bertrand G, Noblet J (2008) British Journal of
Nutrition 100, 1315-1324.
Le Goff G, Noblet J (2001) Journal of Animal Science 79, 2418-2427.
Le Goff G, Le Groumellec L, van Milgen J, Noblet J (2002) British Journal of Nutrition 87,
325-335.
Noblet J, Fortune H, Dupire C, Dubois S, 1993. Animal Feed Science and Technology 42,
131-149.
Noblet J, Shi XS, Dubois S (1993). British Journal of Nutrition 70, 407-419.
Noblet J, Fortune H, Shi XS, Dubois S (1994) Journal of Animal Science 72, 344-354.
Noblet J, Shi XS, Dubois S (1994) Journal of Animal Science 72, 648-657.
Noblet J, Karege C, Dubois S, van Milgen J (1999) Journal of Animal Science 77, 1208-1216
Noblet J, Le Goff G (2001) Animal Feed Science and Technology 90, 35-52.
Noblet J, Bontems V, Tran G (2003) Productions Animales 16, 197-210.
Noblet J, van Milgen J, Carré B, Dimon P, Dubois S, Rademacher M, van Cauwenberghe S
(2003) Progress in Research on Energy and Protein Metabolism. [Souffrant WB, Metges
CC Eds.] EAAP Publication n°109, 205-208, Wageningen Academic Publishers,
Wageningen.
Noblet J, van Milgen J (2004) Journal of Animal Science 82 Number 13, E. Suppl. E229E238.
Noblet J, Sève B, Jondreville C (2004) Tables of composition and nutritional value of feed
materials: pigs, poultry, cattle, sheep, goats, rabbits, horses, fish, [Sauvant D, Perez JM,
34
Aust. Poult. Sci. Symp. 2010...21
Tran G Eds]. Wageningen Academic Publishers, Wageningen and INRA Editions,
Versailles, 25-35.
Noblet J, 2006. Recent Advances in Animal Nutrition 2005. [Garnsworthy PC, Wiseman J
Eds]. Nottingham University Press, Nottingham. 1-26
Noblet J, Dubois S, van Milgen J, Warpechowski M, Le Bellego L, Carré B (2007)
Proceedings, 7èmes Journées de la Recherche Avicole, 141-144
Noblet J, Warpechowski M, Dubois S, van Milgen J, Carré B (2009) Proceedings 8èmes
Journées de la Recherche Avicole, 177-181
Pirgozliev V, Rose SP (1999) World's Poultry Science Journal 55, 23-36
Sauvant D, Perez JM, Tran G (2004) Tables of composition and nutritional value of feed
materials: pigs, poultry, cattle, sheep, goats, rabbits, horses, fish. Wageningen Academic
Publishers, Wageningen and INRA Editions, Versailles.
Schiemann R, Nehring K, Hoffmann L, Jentsch W, Chudy A (1972) Energetische
Futterbevertung und Energienormen. VEB Deutscher Landwirtschatsverlag, Berlin.
Van Milgen J, Noblet J, Dubois S, Bernier JF (1997) British Journal of Nutrition 78, 397-410
Van Milgen J, Noblet J, Dubois S (2001) Journal of Nutrition 131, 1309-1318.
Van Milgen J, Noblet J, Dubois S, Carré B, Juin H (2001) Proceedings, International Animal
Agriculture and Food Science Conference, Poultry Science 80 (Supplement 1), 170
35
Aust. Poult. Sci. Symp. 2010...21
AN EFFECTIVE ALTERNATIVE TO THE METABOLISABLE ENERGY SYSTEM
R. M. GOUS 1
Summary
There are several possible approaches to the development of a net energy system. All seek to
provide a way of stating both how much energy is yielded by a given feed and how much of
this energy a given animal needs to sustain some stated level of performance. Although the
details of the systems which emerge from these approaches differ in detail they are all
essentially ways of predicting the heat increment of feeding. In this paper the effective energy
(EE) system of Emmans (1994), which is equivalent to a net energy system, is presented. The
theory on which the system is based is outlined, the requirements for implementing the system
are described, and the value of the system in terms of simulation modelling is discussed.
Because the EE contents of feed ingredients can be calculated within a feed
formulation program, the system is relatively simple to implement and feeds can be
formulated on an EE basis. Heat production by the bird as a result of consuming a given
amount of a given feed is calculated as the difference between the ME and EE intakes.
Many advantages result from being able to calculate this heat output, and some of these are
given in the paper.
If a change to a net energy system is justified in poultry nutrition then adoption of the
EE scale seems to provide the most readily available alternative.
I. INTRODUCTION
The purpose of an energy system is to describe the energy available in the feed for
maintenance and growth, and for a given level of performance. The metabolisable energy
(ME) system provides information on the amount of energy yielded to a given bird or animal
by a given feed but does not account for the heat increment (HI) of feeding, i.e. it cannot
distinguish between two feeds with the same ME content but differing in their chemical
composition such that the increase in heat production from feeding differs between the two.
The original definition of net energy (NE = ME – HI) proposed by Armsby and Fries
(1915) saw HI as a characteristic that was independent of the level of feeding, and thus as a
characteristic of an ingredient or a feed that could be tabulated, in the same way as is ME.
Similarly, NE could be tabulated also, if this view were accepted. However, this is a false
assumption, as convincing evidence has accumulated demonstrating that HI differs with plane
of nutrition (e.g. Forbes et al., 1928; 1930). Emmans (1984) addressed the problem of
predicting the HI of feeding by first accounting for the fasting heat production, where no
positive retentions of protein or lipid result from the feeding of a given feed, and then
accounting for the extra heat that is produced when these retentions are positive. In his
effective energy (EE) system, which applies to both single-stomached and ruminant animals,
the heat increment of feeding is considered to be linearly related to five measurable quantities
associated with the animal. The EE yielded by a feed or feed ingredient can be estimated
from the ME, the faecal organic matter content and the amounts of digestible crude protein
and lipid. According to Emmans (1994), as the EE values ‘can be tabulated for ingredients,
and are additive to the extent that ME values are additive, they can be used to formulate diets
1
Animal and Poultry Science, University of KwaZulu-Natal, South Africa.
[email protected]
36
Aust. Poult. Sci. Symp. 2010...21
using linear programming’. For these purposes, and for describing animal performance, the
effective energy scale is equivalent to a net energy scale.
In this paper the theory on which the EE system is based is briefly described, the
requirements for implementing the system are outlined, and the value of the system in terms
of simulation modelling are discussed. Some examples are given of the use of the EE system
in describing the HI of feeds for broiler chickens.
II. DESCRIBING THE ENERGY CONTENT OF FEED INGREDIENTS
There are several possible approaches to the development of a net energy system. All seek to
provide a way of stating both how much energy is yielded by a given feed and how much of
this energy a given animal needs to sustain some stated level of performance. Although the
details of the systems which emerge from these approaches differ in detail they are all
essentially ways of predicting the heat increment of feeding.
A general equation for predicting heat increment has been proposed by Emmans (1994)
in which he makes use of the following facts:
(i) Work is done, and hence heat produced, in the catabolism of protein.
(ii) Work is done, and hence heat produced, when protein and lipid are stored in the body.
However, for fat deposition the work done is not directly proportional to the heat of combustion
of the products stored. Less heat is produced when lipid energy in the body is derived from
dietary fat as opposed to energy from the conversion of excess dietary protein (see Table 1).
(iii) More energy is needed to handle bulky feeds than highly digestible feeds due to the higher
faecal organic matter content in bulky feeds.
(iv) Heat is produced when feed is fermented by ruminant animals to produce methane. This
does not apply to simple stomached animals.
These work functions are summarised in Table 1, in which the units of each function are
defined as is the work done per unit.
Table 1.
Definitions of units of work functions (after Emmans, 1994)
Function
Excretion
Fermentation
Defecation
Growth
Fattening
Unit of function
1 kg urinary nitrogen (UN)
1 MJ Methane (MTHE)
1 kg faecal organic matter (FOM)
1 kg protein retention (PR)
1 kg lipid retention (LR)
Symbol
wu
wm
wd
wp
wl
wll
Work, MJ/Unit
30.6 MJ/kg UN
0.675 MJ/MJ methane
3.86 MJ/kg FOM
34.2 MJ/kg positive PR
16.3 MJ/kg LR (or)
4.3 MJ/kg LR1
1
Where feed lipid is used directly for lipid retention by monogastrics
The equation to predict heat increment has five parameters, and these are based on two
propositions:
(i) Apart from maintenance, which includes some level of activity (given), work is done
in the immature animal for only five functions, namely, excretion, fermentation, defecation,
growth and fattening.
(ii) The amount of work, measured as ME, needed for a unit of each function is constant
across animals and diets.
HI = wu (UN - FUN) + wp.PR + wl.LR + wd.FOM + wm.MTHE MJ/d
37
Aust. Poult. Sci. Symp. 2010...21
where FUN is fasting urinary nitrogen, kg/d. This is the urinary nitrogen the animal would have
produced had it not been fed. In monogastric animals MTHE = 0.
Allowance has to be made, in monogastrics, for feed lipid that is used directly for lipid
retention. Of the digested lipid from the feed, DL kg/d, a proportion z is retained with the
production of wll MJ/kg of heat (see Table 1). The lipid retained, other than from dietary lipid
is (LR – z DL) kg/d and the heat produced is wl MJ/kg. On the basis of pig and poultry
experiments analysed by Emmans (1994) he suggests that ‘suitable average values may be z =
0.3 for poultry and z = 1 for pigs’.
The total heat production of lipid retention H (LR) is thus:
H (LR) = wll. z DL + wl (LR - z DL) MJ/d = wl. LR - z DL (wl - wll) MJ/d.
The EE content of a feed or feed ingredient for monogastric animals can thus be defined as
EE = ME - wd. FOM – 0.16wu. DCP + z DL (wl - wll) MJ/kg
or
EE = MEn - 3.8 FOM - 4.67 DCP + 12 z DL MJ/kg
where ME is corrected to PR = 0 and z = 0.3.
The variable z, which may vary from zero (when LR = 0) to unity, introduces an
‘animal’ factor into the equation to calculate the EE content of a feed, with all other coefficients
in the equation remaining constant across ages, physiological states and even species (Emmans,
1994). Noblet et al. (1994) made the point that ‘One diet can theoretically be given several NE
values according to the type of production for which it is used’ and suggested that the
digestibility coefficients of nutrients are influenced by such factors as body weight,
physiological state, feeding level and ‘interaction phenomena’. This leads to the unwieldy use
of different NE values for the same ingredient depending on whether the feed is to be used for
maintenance only, or for growth, fattening or reproduction. The value, z, in the EE system
would appear to have the potential to achieve the same goal.
The effect on EE of a change in the value of z is seen in Table 2 for a range of feed
ingredients. For those containing high levels of digestible crude lipid (DCL) the increase in EE
content is predictably higher than where the ingredient is low in DCL. The resultant effect of an
increase in z would in all cases be to reduce the heat increment of feeding, the effect being
greater with some ingredients than with others.
Table 2.
EE contents of some feed ingredients for three values of z, and the rate of change
in z
Wheat bran
Soybean oilcake
Sunflower oilcake
Fishmeal
Gluten 60
Poultry byprod. meal
Soya full fat
Sorghum
Maize
Sunflower oil
z = 0.3
z = 0.6
z=1
3.74
5.52
5.24
10.19
11.63
10.01
12.14
12.92
13.06
40.75
3.84
5.54
5.27
10.5
11.72
10.44
12.7
13.01
13.18
43.99
3.98
5.58
5.32
10.92
11.84
11.01
13.44
13.13
13.35
48.27
38
Regression
coefficient
0.343
0.086
0.115
1.043
0.300
1.428
1.857
0.300
0.415
10.740
Aust. Poult. Sci. Symp. 2010...21
III. COMPARING THE EE AND ME CONTENTS OF FEED INGREDIENTS
De Groote (1974) used a simplified NE system to differentiate feed ingredients into categories
depending on their contents of crude protein, crude fat and starch + sugar, applying
coefficients of 0.60, 0.90 and 0.75 to these components, which reflected differences in the
efficiency with which the ME of these components is utilised for growth. Protein-containing
ingredients had an NE:ME ratio averaging 0.63, cereals had values around 0.73 and for fats
and oils the ratio was 0.90. Using a similar comparative approach, ten commonly used feed
ingredients have been classified, in Table 3, according to their EE:ME ratio. Because the EE
yielded to a monogastric animal by a feed ingredient will depend on the ME and the relative
proportions of faecal organic matter and digestible crude protein and lipid, the EE:ME ratios
differ to a greater extent between similar ingredients than did De Groote’s NE:ME values. For
example, fishmeal would be classified as a protein-containing ingredient and be given an
NE:ME value around 0.63, but because of the high oil content the EE:ME ratio is considerably
higher than this.
Table 3.
Comparison of EE and ME contents of some feed ingredients (using a value of
z = 0.3)
Ingredient
Wheat bran
Soybean oilcake
Sunflower oilcake
Fishmeal
Gluten 60
Poultry byprod. meal
Soya full fat
Sorghum
Maize
Sunflower oil
Protein
g/kg
Fat
g/kg
Fibre
g/kg
FOM
g/kg
ME
MJ/kg
EE
MJ/kg
EE:ME
155
460
370
655
620
580
370
95
85
0
40
15
15
100
30
140
185
30
38
990
100
50
180
0
15
25
55
25
22
0
514
364
384
149
98
202
251
97
104
0
6.11
8.74
8.20
13.08
14.65
12.56
14.09
13.61
13.66
37.50
3.74
5.52
5.24
10.19
11.63
10.01
12.14
12.92
13.06
40.75
0.61
0.63
0.64
0.78
0.79
0.80
0.86
0.95
0.96
1.09
Because the difference between the ME and EE contents of a feed ingredient is the
amount of heat produced (above maintenance) in the animal when the ingredient is consumed,
those ingredients at the top of Table 3 will have a higher heat increment than the ingredients
near the bottom of the table. This is a useful way of identifying ingredients that could be used
to reduce the heat load on a broiler during hot weather, for example.
IV. CALCULATING THE EE REQUIREMENT OF GROWING BROILERS
One of the main purposes of an energy system is to be able to predict how much of a given
food will be needed to meet the requirements of a bird for maintenance and growth. Having
described how the energy available in the feed for these purposes can be calculated it now
remains to describe how the requirements for maintenance and growth can be calculated.
a) EE for maintenance
The animal fed at maintenance will be producing heat due to the excretion of UN as, by
definition, no protein is retained. This amount of heat will vary depending on the digestibility
of the protein. Maintenance heat (MH) can be estimated from the current and mature protein
39
Aust. Poult. Sci. Symp. 2010...21
weights of the animal, as described by Emmans and Fisher (1986). They proposed the
concept of a maintenance unit, based on the general scaling rule of Taylor (1970), which is
calculated as Pm-0.27. P, where P is the body protein weight (excluding feathers), and Pm the
mature protein weight of the animal. The energy needed for maintenance (ME) was predicted
by Emmans and Fisher (1986) to be 1.63 MJ/unit day, which is used in the following equation
to calculate MH.
MH = ME. Pm-0.27. P MJ/d
On reasonable estimates of Pm and body protein content the value is consistent with calculated
values of MH for cattle, pigs, sheep and poultry at different values of P/Pm (Emmans, 1994).
The MH as calculated here would include some level of physical activity (Emmans,
1994), and could be modified to account, for example, for a cold environment and the
additional activity associated with the consumption of mash as opposed to pelleted feed
(Jensen et al., 1962).
b) EE for growth and fattening
A feed that leads to positive retentions of protein and lipid will be associated with the
production of FOM and UN, and a portion of HI, relative to MH, caused by the feed will be
associated with the rates at which these latter products are produced (Emmans, 1994). It can be
assumed that all energy retention (ER) is accounted for as protein or lipid, so ER = hp.PR +
hl.LR MJ/d, where PR and LR are the rates of retention of protein and lipid, kg/d, and hp and hl
are their respective heats of combustion, MJ/kg. The values of hp and hl of 23.8 and 39.6 MJ/kg
respectively are assumed to be chemical constants.
The quantity of effective energy required (EERQ) by an animal to support given values
of MH, PR and LR can therefore be defined as:
EERQ = MH + PR ((hp – a) + (wp - 0.16 wu)) + LR (hl + wl) MJ/d
where a = 5.63 kJ/g, which is applied to correct the classical ME (MEc) to N-corrected ME
(MEn) using the equation MEn (kJ/d) = MEc – a (6.25 NR), where NR is N retention (g/d). After
substituting the values of the constants into the above equation this reduces to:
EERQ = MH + 50 PR + 56 LR
In the case of poultry, protein retention is in body and feathers, so PR must include both of these
components. The relevant coefficients apply equally to body and feather protein, making the
calculation of EERQ less complicated than when determining the amino acid requirements for
the growth of body and feather protein.
V. IMPLEMENTING THE EE SYSTEM
a) Calculating the EE of feed ingredients
The implementation of the EE scale in feed formulation is quite straightforward. The
information required to calculate an EE value for each ingredient is AMEn, FOM, DCP and
DCL. Starting with the familiar AMEn scale provides a natural evolution of present practice
and also provides a practical basis for quality control. The quantities DCP and DCL
(apparent faecal digestibilities) are not widely used but reasonable table values are available
(e.g. WPSA, 1989) to provide a starting point. New experimental data can be accumulated at
reasonable cost and it is to be expected that most variation in digestibility will be correlated
40
Aust. Poult. Sci. Symp. 2010...21
with variations in ME. Thus quality control of DCP and DCL may be achieved at the same
time as control of ME data.
FOM may be calculated as OM – (DCP + DCHO + DCL), where OM is organic
matter and DCHO is digestible carbohydrate. If we assume a fixed relationship between
AMEn and digestible CP, CL and CHO (see WPSA, 1989 for an example) then DCHO can
be calculated from the data required for the calculation of EE, and thus the digestibility of
the total OM. Care must be taken to ensure that all unaccounted material in the calculation is
CHO (for example anions in mineral sources will need to be called ‘ash’) but with this
proviso the calculations are accurate.
b) Determining MH, PR and LR
The potential protein growth of an animal can be determined using a growth equation such as
the Gompertz growth curve, and if this is done using the approach suggested by Emmans and
Fisher (1986), it is a simple matter to determine the potential EE requirement for a growing
bird on each day of the growing period. It is more difficult to determine the actual growth of
protein and lipid, as this depends on the interaction between the genotype, the food being
offered and the environment in which the bird is kept. A modelling approach is needed to
take account of all these factors, and such a model is available (EFG Software, 2009). In this
model the EE system is used to determine the heat increment of feeding and the heat
associated with maintenance and growth, which enables the prediction of the actual intake of
a given food by a given broiler in a given state under the given environmental conditions,
and this then leads on to the prediction of the actual growth rates of body and feather protein
and lipid.
As a result of using the EE system the model accurately predicts food intake and
hence MH, PR and LR, as opposed to most simulation models in which food intake is an
input to the model and not an output. It is not possible to optimise the way in which broilers
are fed unless food intake is predicted, i.e. unless it is an output from a simulation model.
c) Using EE in place of ME in feed formulation
Within a given practical scenario the effect of implementing a change from ME to EE will
not, in the short term, have a large effect since in a small sample of feed ingredients EE and
ME will be closely correlated. Over time however small changes in ingredient economic
values will lead to changes in ingredient evaluation and use. If a change to a net energy
system is justified in poultry nutrition then adoption of the EE scale seems to provide the
most readily available alternative.
VI. ADVANTAGES OF USING EE vs. ME IN SIMULATION MODELLING
Because the EE system provides a more accurate estimate of the energy available to the bird
for maintenance and growth than does the ME system, there are many advantages to its use.
Some of these are listed below.
a) Calculating the desired food intake (DFI)
To predict food intake the desired amount of food that an animal would need to consume to
meet its potential requirement for each of the essential nutrients needs to be calculated. The
DFI for lysine (DFIlys) would thus be the daily requirement for lysine for maintenance and
potential growth divided by the amount of digestible lysine in the feed. Similarly, the DFI for
each essential nutrient is calculated, and that resulting in the highest DFI defines the first
limiting nutrient in the feed, and would be the amount of food the animal would need to eat
in order to grow or reproduce at its potential. Basing the energy requirement on ME, and
41
Aust. Poult. Sci. Symp. 2010...21
describing the feed energy content in terms of ME in order to calculate the DFIME, would
introduce many inaccuracies and would thus not produce a valid estimate of requirement.
Calculating the DFIEE on the other hand is a more accurate means of predicting food intake
for the reasons described above.
b) Estimating heat production and loss
The heat produced by the bird in consuming a given feed can be calculated as the difference
between the ME and EE intakes. The larger the difference, the greater will be the heat output
per g of feed eaten. One of the constraints that prevent broilers from consuming the amount
of food necessary to achieve their potential output is the environmental heat demand: being
able to maintain a constant body temperature through losing the heat generated to the
environment. Under conditions in which the bird cannot lose all the heat that would be
generated if the DFI were achieved, the only option open to the bird is to reduce food intake
and hence heat production. Because it is possible, with the EE system, to predict the heat
output of a bird in a given state being fed a given food it is also possible to determine
whether the bird would be able to lose that amount of heat to the environment, and to
determine the extent to which food intake would need to be reduced for thermal balance to
be maintained. In this way, one of the most important constraints to the desired food intake
can be addressed.
Similarly, where the environmental temperature is below the bird’s lower critical
temperature the additional amount of energy required by the bird to remain in thermal
equilibrium can be determined, leading to a more accurate estimate of the energy required by
the bird and hence of actual food intake.
c) Nutritional interventions in alleviating heat stress in broilers
It has been proposed that the adverse effects of high temperature on performance might be
alleviated by dietary manipulation. Feeds containing highly digestible nutrients, minimal
excesses of protein (amino acids), and a high proportion of the carbohydrate energy replaced
with digestible fat energy would have the advantage at high temperatures. By increasing the
EE:ME ratio of feeds, as described above, the heat increment of the feed is minimised, and,
theoretically, a greater amount of this feed could be consumed by the bird at high
temperatures, thereby improving performance.
In principle, to minimise heat increment, the EE content of the feed should be as
close to the ME content as possible. With practical ingredients, the extent to which the heat
increment of the feed may be reduced is quite constrained, and therefore, these strategies
afford only marginal improvements in performance (Gous and Morris, 2005). When
formulating feeds containing ingredients commonly used in the broiler industry (e.g. maize,
wheat, soybean meal) the EE:ME ratios that are feasible range from about 0.86 to 0.91, with
the higher ratio resulting when least-cost formulations are made. When the proportion of
ME as lipid in a feed is increased, using commercially available feed ingredients, or when
reducing the dietary amino acid excess as a means of reducing heat stress, the FOM is
invariably increased unless an inert filler such as sand is offered in the formulation. By
monitoring the effect of formulation changes on the EE:ME ratio it is possible both to
explain the lack of response to such interventions when broilers are kept at high
temperatures, and to predict whether any improvement would be possible, if at all.
Most nutritional strategies that have been proposed as a means of reducing the heat of
digestion in the broiler result in a maximum theoretical saving in metabolic heat production
equal to the effect of lowering the dry bulb temperature in the broiler house by about 1ºC
(Gous and Morris, 2005). None of these strategies is as effective in terms of growth rate,
feed conversion, liveability or carcass quality as reducing the radiant heat load on the birds
42
Aust. Poult. Sci. Symp. 2010...21
by making appropriate modifications to the structure of the broiler house and to the
husbandry practices employed.
The ME system is clearly inferior to a NE system, which allows the heat increment of
feeding to be calculated. This measurement is useful in poultry nutrition, especially in
simulation modelling, as it leads to accurate predictions of food intake in different classes of
poultry. For this reason it is almost obligatory to use a NE system when predicting food
intake. If a change to a NE system is justified when formulating feeds for poultry then
adoption of the EE scale seems to provide the most readily available alternative to ME.
REFERENCES
Armsby HP, Fries JA (1915) Journal of Agricultural Research 3, 435–491.
De Groote G (1974) In: Energy Requirements of Poultry [TR Morris, BM Freeman Eds]
British Poultry Science Ltd, Edinburgh.
EFG Software (2009) www.efgsoftware.net (accessed October 2009).
Emmans GC (1994) British Journal of Nutrition 71, 801-821.
Emmans GC, Fisher C (1986) In: Nutrient Requirements of Poultry and Nutritional
Research, [C Fisher, KN Boorman Eds] Butterworths, London.
Forbes EB, Braman WW, Kriss M (1928) Journal of Agricultural Research 37, 253–300.
Forbes EB, Braman WW, Kriss M (1930) Journal of Agricultural Research 40, 37–78.
Gous RM, Morris TR (2005) World’s Poultry Science Journal, 61, 463–466.
Jensen LS, Merrill LH, Reddy CV, McGinnis J (1962) Poultry Science 41, 1414-1419.
Noblet J, Fortune H, Shi XS, Dubois S (1994) Journal of Animal Science 72, 344-354.
WPSA (1989) European Table of Energy Values for Poultry Feedstuffs. Working Group 2,
European Federation. Wageningen.
43
Aust. Poult. Sci. Symp. 2010...21
ENERGY IN POULTRY DIETS: ADJUSTED AME OR NET ENERGY
J D. VAN DER KLIS 1, C. KWAKERNAAK1, A. JANSMAN 2 and M. BLOK 3
Summary
Energy evaluation in poultry is generally based on metabolisable energy systems. For many
decades it has been debated whether or not the development of a net energy system would
enhance the prediction of energy partitioning in meat- and egg-producing poultry. Based on
the extracaloric value of fat in laying hens and a potential lower energetic efficiency of
protein utilisation, several modified ME systems were developed and used in commercial
poultry nutrition. Recently, CVB initiated the development of an ATP-based NE system for
broilers. Validation of this energy system did not show added value over the current modified
AME system for broilers. Therefore, most probably frequent updates in nutrient digestibilities
in feedstuffs would be of higher commercial relevance, as well as target animal dedicated
digestibility data.
I. INTRODUCTION
Feed costs are the main costs for poultry production, of which 70 to 75% are related to dietary
energy. To be able to reduce feed costs while maintaining bird performance, dietary energy
values should be an accurate representation of the utilisable energy for both meat- and eggproducing birds in their successive production phases. This not only refers to the energy
evaluation system used, but it should be realised that digestibilities of the energy delivering
nutrients varies among bird species and age. Generally, energy evaluation systems in poultry
are based on metabolisable energy (apparent metabolisable energy: AME, or true
matabolisable energy: TME), being calculated from digestible protein, digestible fat and
digestible carbohydrates. This carbohydrate fraction might be split into starch, sugars and non
starch polysaccharides.
Discussions are ongoing whether adjustment of AME or TME systems would have
added value for poultry, taking the efficiency of utilisation of digested nutrients into account.
Does it improve the predictability of bird performance and/or reduce feed costs? Furthermore,
the potential of a net energy (NE) system in poultry analogous to swine needs to be evaluated,
as it has been shown in swine nutrition that feed cost savings can be as high as 4.00 to
4.50 €/metric ton when a shift is made from a ME to a NE system, without any negative
impact on production performance. The latter might even be improved due to lower dietary
crude protein levels, which are generally linked to the introduction of a NE system. Of course
feed cost savings depend on the feedstuffs available in different regions in the world and on
the maximum dietary feedstuff inclusion levels for different animal species or categories. In
principle, a NE system would have added value in poultry like in pigs, however, results in
poultry are less conclusive. Differences between ME and NE in pigs are based on a higher
efficiency of utilisation of energy from fat and a lower efficiency of utilisation of energy from
protein compared to starch. In laying hens the extra caloric value of fat has been demonstrated
(eg. Scheele et al., 1985 as cited by Van der Klis and Fledderus, 2007) and is well-accepted,
1
Schothorst Feed Research, PO Box 533, 8200 AM Lelystad, The Netherlands
Animal Sciences Group of Wageningen UR, Lelystad, The Netherlands
3
Centraal Veevoeder Bureau (CVB), Den Haag, The Netherlands
2
44
Aust. Poult. Sci. Symp. 2010...21
but results in broilers are still contradictory. Nevertheless, in time several modified AME
systems have been developed that take the efficiency of nutrient utilisation into account, for
example:
(i) Effective Energy by Emmans (1994).
(ii)for egg-producing poultry using a 15% extra caloric value of digested fat and
for meat-type poultry reducing the energy value of digested protein by approx. 15%
(both modifications being used in commercial poultry feeding in the Netherlands
(CVB, 2005).
Recently, the CVB in the Netherlands initiated the development of an ATP-based NE
system for broilers. The added value of this new system over the currently used AMEbroiler
was evaluated. In 2009, Schothorst Feed Research launched a new feeding table for laying
hens, based on digestibility experiments with highly productive layers. In the current paper
results are presented.
II. MODIFIED AME SYSTEM IN LAYING HEN DIETS
The basis of all energy evaluation systems lies in accurate nutrient digestibilities in feedstuffs.
Until now, feeding values determined in adult roosters are used world-wide to calculate the
dietary energy value for highly productive layer diets. Recently, nutrient digestibilities and
AME values were determined using white LSL laying hens in peak production according to
the European reference assay for poultry (Bourdillon et al, 1990), based on a three-day
droppings collection period. Nineteen feedstuffs were evaluated so far and results were
compared with current tabulated values from restrictedly fed adult roosters (CVB, 2005).
Results are given in Figure 1.
140
120
100
Change (%)
80
60
D.CP
40
D.CFAT
D.NFE
20
0
-20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
-40
-60
Figure 1.
Nutrient digestibilities (digestible crude protein (D.CP), digestible crude fat
(D.CFAT) and digestible nitrogen free extract (D.NFE)) in feedstuffs using
laying hens in peak production. Data are presented relative to current tabulated
values (i.e. restrictedly fed adult roosters) as a reference (CVB, 2005).
45
Aust. Poult. Sci. Symp. 2010...21
It is shown in Figure 1 that especially the digestibility of carbohydrates was higher in
ad libitum fed laying hens compared to restrictedly fed roosters. Differences in nutrient
digestibilities resulted in absolute AME values in feedstuffs (as measured) that were -40 to
530 kcal/kg higher than current tabulated values (CVB, 2005). Although test feedstuffs in
Figure 1 were not simultaneously evaluated in restrictedly fed roosters, it was concluded
based on the large differences between laying hens and adult roosters that the latter have
limited value to determine the energy value of feedstuffs for producing laying hens.
Schothorst Feed Research therefore introduced in 2009 a new feedstuff table for layers.
Currently, validation trials are being done and a new trial was started to evaluate twenty new
feedstuffs and/or different qualities of feedstuffs already tested.
In the Netherlands a modified AME system for laying hens is being used, taking a
higher efficiency of energy utilisation for fat compared to starch into account, as fatty acids
and monoglycerides can be used as such to synthesize fat in body and eggs. Scheele et al.
(1985) estimated the energetic efficiency of fat utilisation in comparison to starch using ad
libitum fed laying hens from 24 to 36 weeks of age. Two sets of diets contained either 11.5 or
12.0 MJ AMEN/kg, each with three levels of dietary fat (3, 6 and 9%) that was isocalorically
exchanged for carbohydrates and ash. Dietary AME, nutrient digestibilities and energy
retention in body and eggs were measured. Results are given in Figure 2.
Figure 2.
The energy retention (in body and egg) in laying hens fed diets differing in
AMEN content (11.5 MJ/kg (●) and 1 2.0 MJ/kg (○)), each containing three
levels of dietary fat which were isocalorically exchanged for carbohydrates.
It is shown in Figure 2 that energy retention in laying hens increases with isocaloric
dietary fat inclusion, indicating a higher energetic efficiency of fat utilisation over
carbohydrates. From this experiment it was calculated that the extracaloric value of fat for
layers was 15% to 19%. The CVB uses a 15% extra caloric value of fat in the Dutch modified
AME system for layers to take this higher efficiency into account. In laying hens no
adjustment has been made for a lower efficiency of protein utilisation, except for the energy
correction for zero nitrogen balance, assuming that all nitrogen from feed will be excreted as
uric acid.
46
Aust. Poult. Sci. Symp. 2010...21
III. DEVELOPMENT OF AN ATP-BASED NET ENERGY SYSTEM
In 2004 a NE formulae was derived based on the ATP yield of carbohydrates, amino acids,
glycerol and fatty acids and volatile fatty acids (Jansman et al. 2004). The results of their
calculations of the ATP yield of nutrients is given as mol ATP/g nutrient and recalculated to
kJ/g in Table 1.
Table 1.
The ATP yield and ATP energy per g nutrient (absolute and relative to starch).
mol ATP/g
Glucose
Starch
Protein1
Fat2
Volatile fatty acids
Fermentable
carbohydrates
0.211
0.235
0.194
0.508
0.212
0.165
ATP energy
kJ/g
10.56
11.73
9.70
25.31
10.60
8.21
Relative to
starch
90
100
83
216
90
70
1
Based on the composition of body protein; 2Based on soya oil
From these coefficients the following equation to calculate the NEATP in feedstuffs
was derived:
NEATP= 9.7 x dig. true protein+ 26.1 x dig. crude fat+ 11.7 x dig. starch +
10.6 x dig. sugars + 8.2 x dig. nitrogen free residue [1]
Subsequently, an equation to calculate the NEATP requirement (kJ/day) was derived
from comparative slaughter experiments based on regression analyses between the NEATP
intake and protein and fat deposition, using male and female broilers from ten different pure
and commercial lines (unpublished data), being:
NEATP (req)= 278 kJ NEATP/BW0.75/day x BW0.75+ 3.058 x energy in protein deposition (kJ
NEATP/day) +1.053 x energy in fat deposition (kJ NEATP/day) [2]
Based on Equation 1 the NEATP content of feedstuffs has been calculated and
compared to the AME equations, using maize as a reference (Table 2). It is suggested in Table
2 that the relative NEATP is of protein rich feedstuffs will be similar to the currently used
AMEbroiler.
Table 2. Relative feeding values of vegetable feedstuffs, using corn as a reference.
Maize
Wheat
Soybean meal
Sunflower seed meal
AMEbroiler1
(growing)
100
91
59
43
AME poultry
(adult)
100
94
67
48
NEATP
100
84
58
42
1
AME broiler is based on an approx. 15% lowered energy value of digestible protein.
The NEATP system has been validated in Ross 308 broilers based on a set of five
diets with a fixed AMEbroiler (12.2 MJ/kg) and variable NEATP (7.7 to 8.3 MJ/kg) and five diets
with a fixed NEATP (8.0 MJ/kg) and variable AMEbroiler (11.8 to 12.6 MJ/kg) to test whether
47
Aust. Poult. Sci. Symp. 2010...21
the new NEATP equation has added value over the current AMEbroiler equation to predict the
energy retention in broilers. Broilers were fed twice a day from day 8 onwards, and daily feed
allowance from was standardised based on pairwise feeding. Nutrient digestibilities were
measured from 13-17 days of age and from 27-31 days of age and energy retention was
determined using the comparative slaughter technique from 8 to 24 days of age and from 24
to 36 days of age. Average feed intake from 8 to 36 days was 82.7 g/day (approx. 75% of ad
libitum intake for Ross 308 in this period). Body weight gain during the experimental period
was 1557 g, resulting in an average feed conversion efficiency of 1.491.
The energy retention from 8 to 36 days of age is given in Figure 3. AME and NE were
related in this experiment, because the current modified AMEbroiler equation was used.
Approx. 65.0% of the variation in energy deposition was explained by AMEbroiler intake and
only 60.3% by NEATP intake. It was therefore concluded that in this experiment NEATP had no
added value over AMEbroiler (Jansman and Van Diepen, 2008), in which broilers were
restricted fed.
16.5
Energy retention (MJ)
16.0
15.5
15.0
14.5
Trt. I-V
14.0
Trt. III, VI-IX
13.5
13.0
12.5
12.0
16.0
16.5
17.0
17.5
18.0
18.5
19.0
19.5
20.0
20.5
NE intake (MJ NEATP)
Figure 3.
The Energy retention (MJ) in broilers from 8 to 36 days of age as affected by
NEATP intake. Diets I to V were formulated on variable AMEbroiler content and
diet III and VI to IX on variable NEATP content (both in MJ/kg).
IV. NET ENERGY VERSUS METABOLISABLE ENERGY SYSTEM
Pirgozliev and Rose (1999) quantitatively reviewed the relationship between dietary AME,
NE and Effective Energy on the one hand and energy deposition (or net energy for
production: NEp) on the other hand, using a rather old dataset of 40 feedstuffs varying in
AME value from 8.0 to 16.5 MJ/kg. NEp was measured as energy deposition in young
growing chickens depositing mostly lean meat (Fraps 1946, as cited by Pirgozliev and Rose,
1999). Based on regression analyses, Pirgozliev and Rose (1999) concluded that the AME
value (calculated as 22.4 x D.CP+ 39.2 x D.CFAT + 17.2 D.NFE (in MJ/kg)) overestimated
NEp for high protein feedstuffs. The regression equations obtained (after we omitted two
feedstuffs (dried butter milk and dried skimmed milk) with a high leverage from their dataset)
were:
NEp= 0.711 x AME - 0.598 (r2=0.92) for cereal and cereal by-products and
NEp= 0.545 x AME + 0.843 (r2=0.81) for high protein feedstuffs
48
Aust. Poult. Sci. Symp. 2010...21
These authors concluded that correcting the AME for crude protein content gave an
equally accurate prediction of NEp as Effective Energy according to Emmans (1994) and Net
Energy according to Hoffmann and Schiemann (1980) and De Groote (1974) as cited by
Pirgozliev and Rose (1999). This is more or less as expected as the dietary AME value was
not yet corrected for zero nitrogen balance, using an energy value of 22.4 instead of 18.0
MJ/kg digested CP. Based on their regression analyses, the energy value of digested protein
most probably needs to be corrected below 18.0 MJ/kg D.CP. Pirgozliev and Rose (1999) did
not show any proof for an extracaloric value of fat in broilers, which can be attributed to the
lack of feedstuffs with higher fat levels in their dataset. Nitsan et al. (1997) did show an
extracaloric effect of 2.6% soybean oil (plus wheat bran) isocalorically exchanged for maize
meal in broilers at 12.1 (comparing 2.6 and 5.3% fat) and 13.1 MJ/kg (comparing 6.9 to 9.5%
fat). At the low dietary AME level heat production was reduced by 14% due to soybean oil
inclusion, whereas effects at the high AME level were only small. Based on their calculations
the extracaloric effect was 17% in the low, and 3% in the high AME diet. They suggested that
the extracaloric effect is related to bird age (no effect expected in young birds having a low fat
digestibility), dietary fat level and source: An excess of unsaturated fatty acids might even
stimulate heat production. More recently, Noblet et al. (2009) were not able to show any
effect of the dietary fat content in broiler diets on their heat production. The NE/ME ratio in
their trial was approx. 75% irrespective of the dietary fat content (2.9% versus 9.6%). Noblet
et al. (2007) did not demonstrate any increase in heat production either using high protein
diets in 4-5 week old broilers. The NE/ME ratio was remained 68% although the dietary crude
protein level was increased from 22.5 to 27.3%. It can therefore be questioned whether a net
energy system would have any significance to broilers, despite a few references that might
indicate some benefits of a NE system. A (modified) AME system most probably would be
adequate to describe energy partitioning in broilers.
REFERENCES
Bourdillon AB, Carré L, Conan M, Francesch M, Fuentes G, Huyghebaert WMMA, Janssen
B, Leclerq M, Lessire J, McNab M, Rigoni, Wiseman J (1990) British Poultry Science
31, 567-676.
Emmans GC (1994) British Journal of Nutrition 71, 801-821.
Jansman AJM, Kwakernaak C, van der Klis JD, de Bree J, Fledderus J,. Brandsma GG, Blok
MC (2004) A net energy system for broilers (in Dutch). ASG-report 04/5917.
Jansman AJM, van Diepen JTM (2008) Validation of a new net energy system for broilers (in
Dutch). ASG-report 08/.
Nitsan Z, Dvorin A, Zoref Z, Mokady S (1997) British Poultry Science 38, 101-106.
Noblet J, Dubois S, van Milgen J, Warpechowski, M Carré B (2007) Heat production in
broilers is not affected by dietary crude protein. In: Energy and production metabolism
and nutrition (I Ortigues-Marty, N Miraux, W. Brand-Williams Eds.) pp 479-480.
Noblet J, Warpechowski M, Dubois S, Van Milgen J, Carré B (2009) Influence of feed fat
content on metabolic utilization of energy in growing chickens. Proceedings of the 8e
Journées de la Recherche Avicole. Saint-Malo, France.
Pirgozliev V, Rose SP (1999) World’s Poultry Science Journal 55, 23-36.
Scheele CW, van Schagen PJW, van Es AJH (1985) Energy utilisation in layer diets (in
Dutch). COVP report 005.
Van der Klis JD, Fledderus J (2007) Evaluation of raw materials for poultry: What’s up?
Proceedings of the 16th European Symposium on Poultry Nutrition. Strasburg, France.
49
Aust. Poult. Sci. Symp. 2010...21
NON-STARCH POLYSACCHARIDES AND ENZYME APPLICATION INFLUENCE
THE NET ENERGY VALUE OF BROILER DIETS
M. CHOCT 1, A. TUKEI1 and D.J.CADOGAN 2
The apparent metabolisable energy (AME) assay is a default system of energy estimation in
the poultry industry, but the system is not capable of accounting for losses of chemical energy
in the solid, liquid and gaseous excreta or as heat (Pirgozliev and Rose, 1999). Measuring
energy lost to account for heat increment (HI) and excreta volatile fatty acids (VFA), and thus
calculating the net energy (NE) of the diet, should better explain the negative influence of nonstarch polysaccharides (NSP) and benefits of their breakdown by xylanases. Four groups of
day-old mixed broiler birds (Cobb strain), each consisting of 2 randomly selected birds of
similar weights, were selected from 2 different batches. On d 17 birds were transferred to 4
closed circuit respiration chambers (2 per cage) in a climate-controlled room (25 ± 2oC) under
continuous fluorescent lighting. On d 4 and 5, the birds were deprived of feed to establish their
basal metabolism and fasting heat (FH). During d 6-9, they were fed a wheat-based diet
containing 4% soluble NSP, and that diet with a xylanase product (250 mg Ronozyme WX),
followed by collection of excreta collection for determination of AME, as well as measures for
respiratory quotient (RQ), heat production (HP) and HI during d 10-14. On d 14 after the last
collection, all the birds were sacrificed, and intestinal contents were collected for digesta
viscosity (data not shown) and VFA analysis.
Table 1.
The effect of soluble non-starch polysaccharides (NSP) and xylanase
supplementation on energy utilisation
Measure
RQ
HP (MJ/kg wt/day)
HI (HP-FH) (MJ/kg wt/day)
HI (MJ/kg feed)
AME (MJ/kg feed)
AME-HP (NE: MJ/kg feed)
Wheat + NSP
(- enzyme)
Mean
SE
0.995
0.014
0.91
0.013
0.20
0.01
1.28
0.11
10.5
0.06
4.7
0.23
Wheat + NSP
(+ enzyme)
Mean
SE
1.006
0.02
0.81
0.02
0.13
0.01
1.17
0.09
14.8
0.07
7.5
0.16
P
value
0.772
0.103
0.004
0.402
0.003
0.008
RQ: respiratory quotient; HP: heat production; HI: heat increment; FH: fasting heat; AME: apparent
metabolisable energy (MJ/kg); NE: net energy (MJ/kg); wt: body weight; SE: standard error.
Enzyme supplementation did not affect RQ HP, but it numerically lowered HP by
11.0%. Enzyme reduced (P<0.01) HI based on per kg bodyweight/day but not on the basis of
MJ/kg feed. AME and NE increased (P< 0.01) with enzyme supplementation (Table 1).On a
4-day average basis, energy loss as VFA was 101.5kJ/chamber/d for birds fed the control diet
compared with 34.3kJ/chamber/d for those fed enzyme-supplemented diet. Overall, xylanase
reduced HP by 11% and decreased energy lost to excreta VFA by 66.2% which resulted in
enzyme increasing NE (37%) proportionally more than AME (29.1%). This further supports
the argument to establish an NE-based feed formulation system.
Pirgozliev V, Rose SP (1999) Brit. Poult. Sci. 55, 24-36
1
The University of New England, School of Environmental and Rural Science, Armidale, NSW 2351
2
Feedworks, Romsey Vic 3434
50
Aust. Poult. Sci. Symp. 2010...21
UPDATE ON NEAR INFRARED REFLECTANCE ANALYSIS OF GRAINS TO
ESTIMATE NUTRITIONAL VALUE FOR CHICKENS
J.L. BLACK 1, R.J. HUGHES 2, M.S. GEIER2, S.G. NIELSEN 3, A.M. TREDREA 4 and
P.C FLINN 5
Summary
Previously reported near infrared (NIR) calibrations for predicting the apparent metabolisable
energy (AME) content (MJ/kg as fed) and AME Intake Index of cereal grains have been
updated with results from an additional 55 grains. The new calibrations have improved greatly
the ability to predict values for unknown grains and the accuracy of prediction. The new
calibrations are available to industry under license through the Pork CRC project, AusScan.
I. INTRODUCTION
The apparent metabolisable energy (AME) content (MJ/kg) of cereal grains and the amount
eaten by broiler chickens when the grains are incorporated into diets are major determinants
of the efficiency of feed use, time taken for birds to reach market weight and profitability of
the broiler industry. The AME content of wheat, barley and triticale has been shown to range
between samples by over 3 MJ/kg (Scott, 2004; Black et al., 2005). Similarly, the intake of
energy (MJ/day) by broilers varies by approximately 34% when different samples of wheat
are incorporated into diets (Black, 2008). This large variation in the energy value of cereal
grains means that a rapid method is required for measuring AME and relative AME intake for
individual batches of grain if profitability of broiler enterprises is to be maximised.
Black et al. (2009) described calibrations based on near infrared reflectance (NIR)
spectroscopy for estimating the AME content and AME Intake Index of cereal grains for
broiler chickens. The calibrations were developed from results obtained in the Premium
Grains for Livestock Program (PGLP; Black, 2008). AME Intake Index (0-100) was used
rather than AME intake (MJ/day) to estimate the relative effect of a grain sample on broiler
feed intake, because AME intake changes each day as a bird grows.
The PGLP calibrations were established from results for just over 100 grain samples
(Black et al., 2009). This number is regarded as being near the minimum needed for reliable
and robust NIR calibrations for predicting accurately values for unknown samples. There
were relatively few weather damaged and ‘pinched’ grains included in the PGLP samples. In
addition, the ability of the calibrations to predict the energy value of unknown samples of
grains was not tested. Consequently, two experiments, one including mainly weather damaged
grains and the other with grains supplied by the Australian broiler industry, have been
conducted to evaluate the ability of the NIR calibrations to predict the energy values for
unknown grains. The results from these experiments were included with the PGLP results to
upgrade the calibrations.
II. EXPERIMENTS
The experimental procedures used in PGLP (Black, 2008) were adopted in two experiments to
measure the AME content (MJ/kg) and AME intake (MJ/day) of cereal grains incorporated
1
John Black Consulting, Warrimoo, NSW 2774
SARDI, Pig and Poultry Production Institute (PPPI), Roseworthy, SA 5371
3
NSW Industry and Investment, Orange, NSW 2800
4
University of Sydney, Plant Breeding Institute, Narrabri, NSW 2390
5
Kelspec Services Pty. Ltd, Dunkeld, VIC 3294
2
51
Aust. Poult. Sci. Symp. 2010...21
into diets for broiler chickens. Experiment 1 included 30 weather damaged wheat (10), barley
(11), triticale (3) and sorghum (6) grains, with a range in screenings content (< 2.0 mm) from
1% to 67%. Experiment 2 contained 25 grains (wheat, 9; barley, 7; triticale, 2; sorghum, 7)
provided by the Australian broiler industry. Each experiment included approximately 30%
additional grains that had been used in previous experiments to provide connectivity for
statistical analysis needed to remove identified variation between experiments. Each grain
sample was formulated into two diets, one containing xylanase and phytase enzymes and the
other without enzymes. The results reported are for grains without enzymes.
A sample of each grain was scanned using a Foss 6500 instrument at the time the grains
were first fed to the birds. Scans were conducted on whole and milled grain samples. Standard
NIR technology was adopted to develop updated calibrations using WINISI software. For
each experiment, the latest existing calibration was first used to predict the values for new
grains. These predicted values were compared with the measured values and then included
with existing values to update the calibrations. Results for calibrations based on whole grain
scans only are presented.
III.
ACCURACY OF PREDICTIONS FOR NEW SAMPLES
The statistics for linear regression equations fitted to NIR predicted and measured values for
AME and AME Intake Index for all grains in experiments 1 and 2 are given in Table 1. The
ability of the calibrations established in PGLP to predict accurately the AME and AME Intake
Index values for the weather damaged grains from experiment 1 was only fair (AME - R2 =
0.63, slope = 0.89; AME Intake Index - R2 = 0.68, slope = 0.93 for whole grain scans). This
result was expected because the grains used to establish the PGLP calibrations contained a
small range in screenings content. When the grains from experiment 1 were included with
PGLP grains in a new calibration, there was a substantial improvement in ability to predict the
values measured for experiment 1 grains, with R2 for the regressions exceeding 0.8. The real
test of the new calibrations was their ability to predict the values for a new set of unknown
grains from experiment 2. The new calibrations predicted values for experiment 2 grains with
R2 values of approximately 0.8, which was a considerable improvement. Again, inclusion of
experiment 2 grains with experiment 1 and PGLP grains in the development of third
generation calibrations showed a further improvement in the ability to predict AME and AME
Intake Index for experiment 2 grains, with R2 values of approximately 0.9. In addition to
improvements in the R2 values, the slope of the regression between measured and predicted
values improved from around 0.9 to near 1.0 and the intercepts were closer to zero. These
values suggest that with the latest calibrations there is little bias in the ability to predict values
for unknown samples.
IV.
IMPROVEMENTS IN CALIBRATIONS
Including results from experiments 1 and 2 with PGLP results produced a modest, but
continuous improvement in the calibrations for predicting the AME content and AME Intake
Index (Table 2). R2 values for the calibration predicting AME content increased with each
additional set of grains from 0.88 to 0.91, the robustness of the calibration (RPD) increased
from 2.40 to 2.67 and the precision of prediction improved with a decrease from 0.48 to 0.45
MJ/kg. There were similar improvements in AME Intake Index calibrations. R2 values
increased from 0.84 to 0.87 and the precision of prediction improved from 4.85 to 4.22 units.
The biggest improvement was in the calibration robustness, with RPD increasing from 1.82 to
2.38. The increase in RPD values for both AME and AME Intake Index indicates that the
chances of obtaining spurious results are much diminished.
52
Aust. Poult. Sci. Symp. 2010...21
Table 1.
Linear regression coefficients for equations fitted to NIR predicted values for
AME and AME Intake Index and measured values for grains in experiments 1
&2
Calibration used for prediction
R2
AME (MJ/day as fed)
Experiment 1
PGLP calibration (1)
0.63
PGLP + Exp 1 calibration (2)
0.81
Experiment 2
PGLP + Exp 1 calibration (2)
0.81
PGLP +Exp 1 + Exp 2 calibration (3)
0.91
AME Intake Index (0-100)
Experiment 1
PGLP calibration (1)
0.68
PGLP + Exp 1 calibration (2)
0.80
Experiment 2
PGLP + Exp 1 calibration (2)
0.79
PGLP + Exp 1 + Exp 2 calibration (3)
0.88
Table 2.
Slope
Intercept
0.89
1.02
1.62
-0.22
1.09
1.05
-1.01
-0.64
0.93
0.96
6.61
2.84
0.95
1.01
6.83
0.25
NIR calibrations information as the number of samples used is increased from
PGLP by adding results from experiments 1 & 2
Calibration
N1
PGLP (1)2
(1) + Exp 1 (2)
(2) + Exp 2 (3)
102
147
180
PGLP (1) 2
(1) + Exp 1 (2)
(2) + Exp 2 (3)
104
147
184
Mean1
SD1
RSQ1
SEC1
AME (MJ/kg as fed)
12.40
1.15
0.88
0.40
12.49
1.16
0.90
0.37
12.61
1.20
0.91
0.37
AME Intake Index (0-100)
64.8
8.84
0.84
3.57
67.1
8.98
0.85
3.54
69.3
10.0
0.87
3.64
1-VR1
SECV1
RPD1
0.82
0.84
0.86
0.48
0.46
0.45
2.40
2.52
2.67
0.70
0.77
0.82
4.85
4.31
4.22
1.82
2.08
2.38
1
N, observations used in calibration; Mean of observations; SD, standard deviation of observations; RSQ, R2 for
predicted and observed relationship; SEC, standard error of calibrations; 1-VR, 1-Variance Ratio or fraction of
variance accounted for when some observations are used for ‘cross validation’; SECV, standard error of cross
validation, RPD, ratio of Prediction to Deviation = SD/SECV an indication of the value of the calibration, < 1.5
calibration unsatisfactory; 1.5–2.0: calibration can distinguish between high & low values; 2.0-2.5: calibration is
quantitative; 2.5-3.0: calibration predictions good; > 3.0: calibration predictions excellent.
2
Values differ slightly from Black et al., (2009) because of calibration revision.
V.
DISCUSSION
Inclusion of additional grain samples in the calibrations has substantially improved the ability
to predict AME content and AME Intake Index of unknown samples. Addition of new grains
increased the R2 for the calibration predicting AME content from 0.88 to 0.91, which is
greater than 0.35 for an AME calibration derived from 94 whole grain scanned wheat samples
developed by Owens et al. (2009). Similarly, the R2 value was only 0.45 for predicting AME
of wheat from a calibration based on milled grain scans of 160 samples by Garnsworthy et al.
(2000).
RPD, which reflects the reliability for predicting values for unknown samples, of the
calibration for predicting AME content of grains also increased from 2.40 to 2.67 as the grains
from the new experiments were included. NIR experts regard calibrations with RPD values
greater than 2.5 as being reliable for predicting values for unknown samples. However,
SECV, which provides an indication of the likely precision of prediction, decreased only from
53
Aust. Poult. Sci. Symp. 2010...21
0.48 to 0.45. This result means that the latest calibration can predict with 95% confidence to
within ± 0.45 MJ/kg as fed of the actual AME value. Whereas the addition of more grains to
the calibration has increased its ability to predict values for a wider range of grains, it has not
greatly increased the precision of the prediction.
The SECV value obtained for the AME content of cereal grains for broiler chickens is
considerably higher than the value of 0.27 MJ/kg obtained for the digestible energy (DE)
content of cereal grains for pigs when PGLP and Pork CRC experiments are combined. The
numbers of grains used to establish the calibrations were similar for both species (180
broilers; 170 pigs), but the average standard error for measured AME values for broilers was
0.21 MJ/kg compared with 0.11 MJ/kg for DE in pigs. The lower variation in measurements
of DE in pigs compared with AME in poultry may be due to the experimental designs and
statistical correction of results because pigs were housed individually and broilers in cages of
five birds. In addition, differences in gut microbial population between birds and variation in
mean retention time of digesta in the gastrointestinal tract of broilers are known to have an
effect on AME values (Torok et al., 2008; Hughes 2008). Nevertheless, there appears to be
substantial opportunity for improving the precision of NIR calibrations for broilers through
the inclusion of results from more grains.
Calibrations for AME Intake Index also improved as the grains from the new experiments
were added to the data from PGLP, with R2 increasing from 0.84 to 0.87 and RPD increasing
from 1.82 to 2.38. The latter value suggests that the calibration is valuable, but could be
improved further with the addition of more grains. The new calibrations are available under
license to the animal and grains industries through the AusScan project of the Pork CRC.
VI.
ACKNOWLEDGEMENTS
Partial funding from the Rural Industries Research and Development Corporation Chicken
Meat Program and the Cooperative Research Centre for an Internationally Competitive Pork
Industry is gratefully acknowledged. The Pork CRC provided grains for experiment 1.
REFERENCES
Black JL (2008) Premium Grains for Livestock Program: Component 1 – Coordination. Final
Report. Grains R&D Corporation, Canberra, Australia.
Black JL, Hughes RJ, Nielsen SG, Tredrea AM, Flinn PC (2009) Proceedings, Australian
Poultry Science Symposium 20, 31-34.
Garnsworthy PC, Wiseman J, Fegeros K (2000) Journal of Agricultural Science, Cambridge,
135: 409-417.
Hughes RJ (2008) British Poultry Science, 49, 716-720.
Owens B, McCann MEE, McCracken KJ, Park RS (2009) British Poultry Science, 50, 103122.
Torok VA, Ophel-Keller K, Loo M, Hughes RJ (2008) Applied and Environmental
Microbiology, 74, 783-791.
Scott TA (2004) Proceedings, Australian Poultry Science Symposium 16, 9-16.
54
Aust. Poult. Sci. Symp. 2010...21
THE PERFORMANCE OF BROILERS OFFERED SORGHUM- COMPARED TO
WHEAT-BASED DIETS: A LARGE SACLE EXPERIMENT
R.A. PEREZ-MALDONADO 1 and H. RODRIGUES1
Summary
Sufficient evidence tended to indicate that at least four factors can negatively influence
broiler performance when offered sorghum-based diets; in particular energy utilisation of
sorghum in young birds. It was proposed that mainly CT would further influence sorghum
grain AME values when consumed by young chicks (0-7 and 7-14 d old). Overall, birds
consuming sorghum-based diets during the starter phase (0-21 d), did not match the
performance of birds offered wheat-based diets. The use of phytase enzymes in sorghumbased diets tended to improve bird performance. However, reducing the obtained AME of
sorghum grains by -0.8 MJ during the 0-21 d period appears to be a practical solution.
I. INTRODUCTION
There are at least four factors linked to reduced broiler performance and carcass quality when
birds are offered sorghum-based diets. These factors are: (i) low cystine (52%) and
tryptophan (63%) digestibility, (ii) the high link between condensed tannin (CT) found in
sorghum and low tryptophan digestibility (r = -0.673), (iii) the high link between sorghum
CT fractions lowering apparent metabolisable energy (AME) of sorghum in young birds and
(iv) the high phytate-P content (76%) of sorghum. Regarding factor (iii), the reduced
sorghum AME value has been calculated to be about -1.0 MJ/Kg DM in younger birds (14-21
d) when compared with AME values obtained in older birds (22-28 d). Normally in Australia,
the AME values obtained with 22-29 d old birds are used to formulate diets for birds up to 21
days of age. Such an AME difference has a significant nutritionally negative outcome
particularly during the starter phase (0-21 d old) where a poor feed conversion ratio (FCR) at
21 d has been observed. This poor energy utilization has been linked (r = 0.704) with total
intakes of CT from sorghum grain during same period. When the sorghum grain AME
adjustment value of -1 MJ was imposed to formulate new sorghum-based broiler diets, an
excellent grower phase (22-42 d) performance and carcass quality were recorded. But during
the starter period (0-21 d), the poor FCR in birds offered the sorghum-based diets with the
adjusted sorghum AME was still present as compared with those on wheat-based diets
(Perez-Maldonado and Rodrigues, 2009). Thus, the present experiment tests the hypothesis
that, the observed detrimental effect of sorghum CT, would further reduce the AME value in
younger chicks, 0-7 and 7-14 d old. Therefore, it is intended to confirm this assumption
including an examination of other feeding alternatives that may provide information on
improving the use of sorghum-based diets for broiler production.
II. MATERIALS AND METHODS
This experiment was conducted with 1800 (900 males and 900 females) Arbor Acres birds
kept in floor pens from 0-42 days. The 64 pen building has artificial lighting, electric heating
brooders with side shutters, cooling fans and sprinklers devices installed and electronically
controlled. Each pen measured 7.5m2 and was covered with wood shavings with feed and
water available ad libitum. Each pen housed 30 chickens until 42 days of age, with a stocking
density of 4 birds/mt2. During the experiment, there was a 23 h/d lighting period from 1-42 d
1
Department of Primary Industries and Fisheries, ARI, Yeerongpilly, Qld 4105, Australia
55
Aust. Poult. Sci. Symp. 2010...21
of age and the temperature was gradually reduced from 29-32 °C in accordance to industry
practice for maximum bird comfort to 26-24 °C by the time they reached 21 days and to 2221 °C (when possible) by the time they reach 42 days.
Day old chicks were received from a commercial hatchery; weighed in groups of 30
birds each and randomly allocated to pens and offered ad libitum the corresponding
experimental starter diets. At day 21, pen feed residues and chick groups from each pen were
weighed, with starter diet replaced with their equivalent grower/finisher dietary treatments
which was offered ad libitum until day 42. Then again feed residue and birds from each pen
were weighed. During the experiment chickens that died, or were culled during the first 72 h,
were replaced by healthy birds and any bird dying thereafter was not replaced. Birds that died
or were culled and not replaced were individually weighed at the time of removal from the
pen and the pen feed residues recorded in affected pens.
The 10 dietary treatment were (1) control wheat-based diet, (2) sorghum H Pioneer
Bonus, (3) sorghum M Pacific MR Buster, with reduced the obtained sorghum AME value by
-0.8MJ in starter phase only to switch to sorghum M during finisher period using its normal
grain AME value; (4) sorghum B Pacific Buster, (5) sorghum E Pacific Buster, with the
obtained sorghum AME reduced by -0.8MJ in the starter phase only to switch to sorghum E
during the finisher period using its normal AME value, (6) commercial Brisbane sorghum,
(7) sorghum K Hylan Lyberty during starter phase and the commercial sorghum during
finisher phase (insufficient sorghum K); (8) control wheat-based diet during starter period
and switch to commercial sorghum during finisher period, (9) commercial sorghum with
phytase added on top and (10) commercial sorghum with phytase added after reducing
available phosphorous from 0.44 to 0.38% in the starter period and from 0.34 to 0.30% in the
finisher period. Ca was reduced from 0.85 to 0.75% and from 0.68 to 0.60% in the starter and
finisher periods, respectively.
Therefore, this trial examined whether the FCR can be improved during the starter
phase (0-21 d) by changing sorghum AME specifications. Two sorghum-based, diets 3 and
diet 5 were formulated with corresponding sorghums M and E in which its determined AME
was lowered by -0.8 MJ/Kg but only during the starter phase. Then the diets used the actual
measured grain AME value during the finisher phase. The responses to these formulations
were then compared with the response to diets containing sorghums formulated on the actual
measured sorghum AME value (Sorghum H, Sorghum B and commercial sorghum).
III. RESULTS AND DISCUSSIONS
The results (Table 1) indicate that during the 0-21 d period birds given treatments 2, 6, and 7
representing diets based on sorghums H, commercial and K, had a similar feed intake (FI) to
birds consuming the wheat control diet. But these sorghums produced significantly (P < 0.05)
poorer live weight gain (LWG) and FCR. Sorghum K of low CT content did not perform as
expected in the present experiment and it was suspected that it had deteriorated before the
diet formulation. Birds on sorghum treatments 3, 4, 5, 9 and 10 representing sorghum diets
M, B, E and commercial sorghums with the phytase enzyme treatments added, exhibited a
higher (P < 0.05) FI with a similar LWG when compared with the control birds, but resulted
in a poorer FCR. It is noteworthy in this study, that birds consuming the sorghums M and E
(diets 3 and 5) which had their AME reduced by -0.8MJ, exhibited a superior (P < 0.05) FCR
among sorghums (1.398 and 1.391 respectively) and were only 3.2% and 2.7 % respectively
less efficient than the control wheat diets (FCR 1.354). During the finisher period (22-42 d),
except for the birds on diets 2 (sorghum H) and 4 (sorghum B), all birds on the sorghumbased diets had a similar (P > 0.05) FI to birds fed the control wheat diet with birds on all
sorghum based diets exhibiting a significantly (P < 0.05) superior LWG and FCR than the
control birds resulting in a superior overall LWG and FCR at 42 days. But the birds
56
Aust. Poult. Sci. Symp. 2010...21
consuming sorghum-based diets 3 and 5 that had their sorghums AME reduced by -0.8MJ
during the starter phase, exhibited the largest LWG (1980 and 1923 g, respectively) of all
treatments, with a similar FCR among sorghums cultivars at 42 d of age. In terms of bird
energy intake and performance, the same diets 3 and 5, with reduced grain AME by -0.8 MJ,
soy oil was added to be iso-caloric (12.5 MJ/kg diet) as with the other treatments assessed.
The results show (Figure 1a) that during the 0-21 d period, birds offered these reduced
sorghum energy based diets had a similar grain energy intake, producing similar LWG as
birds in the wheat diets. Similarly, the FCR of birds fed on the sorghums based diets 3 and 5
(sorghums M and E) improved when compared with birds offered other sorghum diets
(Figure 1b). It would appear that the addition of oil to these two starter diets was required, to
compensate for the lower energy utilisation of the sorghum grain by young birds. Our results
agree with Douglas et al. (1990) who indicated the need for adding animal fat to typical
broiler diets, to compensate for the lower MEn of CT sorghum grains. It was suggested that
the low energy efficiency with CT sorghum grains is caused, in part, by the cross-links
between protein in sorghum grain, which decreases the digestibility of the protein and of the
starch embedded in it. Another reduction on energy efficiency is due to undigested tanninprotein complexes. In the present experiment, the oil content of the sorghum diets 3 and 5
was raised 2.9% and 1.1%, respectively, to compensate for the -0.8 MJ AME reduction
values in their respective sorghums during the diet formulation of the starter period.
Therefore, both due to age of bird and the highly negative CT effect seen, it is necessary for
this grain AME adjustment value to be applied in sorghum-based broiler diets in order to
improve sorghum efficiency particularly during the early starter period 0-14 d.
(a)
Diet 3
Live weight gain (g) 0-21 d
790
Wheat diet
1.44
780
770
760
750
740
8.0
(b)
1.46
Diet 5
FCR (FI:LWG) 0-21 d
800
1.42
1.40
Diet 3
Diet 5
1.38
Wheat diet
1.36
8.5
9.0
9.5
10.0
1.34
8.0
8.5
Grain AME intake (MJ/0-21 d)
Figure 1.
9.0
9.5
10.0
Grain AME intake (MJ/0-21 d)
Relationship between grain AME intake and live weight gain (a) and AME intake and feed
conversion ratio (FCR) (b) of broiler during the starter period when given sorghum ●)
( or wheat
(○) based diets. Each point is the mean value of six floor pens (30 birds/pen)
During the grower/finisher period, no extra oil was needed in these M and E
sorghums diets which were formulated with the actually determined grain AME value. The
results showed that during the grower/finisher period diets with sorghum grain M and E and
all other sorghum diets produced excellent bird performance (Table 1) with no need to add
extra oil during this period. Future research on sorghum grain to explain the reasons for this
lower energy utilization during the starter period only is warranted.
REFERENCES
Douglas JH, Sullivan TW, Bond PL, Struwe FJ (1990) Poultry Science, 69, 1147-1155.
Perez-Maldonado RA, Rodrigues H.D (2009) RIRDC Publication No 07/ DAQ-326A.
57
Aust. Poult. Sci. Symp. 2010...21
Table 1.
Mean feed intake (FI), live weight gain (LWG), feed conversion ratio (FCR) of broilers in semi-commercial floor pens for starter (021 d) grower/finisher (22-42 d) and overall (0-42 d) periods when fed sorghum-based diets (2005 sorghum harvest) with various
strategies applied or a wheat-based (control) diet
Diet
Treatment
No
1
2
3
4
5
6
7
8
9
10
Age in Days
Wheat control (both starter and grower)
Sorghum H both starter & grower
Sorghum M reduced AME for starter & normal AME- grower
Sorghum B (both starter & grower )
Sorghum E reduced AME for starter & normal AME- grower
Sorghum commercial both starter & grower
Sorghum K for starter & Sorghum Commercial for grower
Wheat for starter & sorghum commercial for grower
Sorghum commercial + Phytase both starter & grower
Sorghum commercial + Phytase -AvP & -Ca starter & grower
LSD (P=0.05)
FI (g/bird)
LWG (g/bird)
FCR (feed:lwg)
0-21
1077b
1085abd
1114cd
1116cd
1110ac
1079ab
1053b
1077b
1117cd
1128c
22-42
3567bc
3464a
3607b
3466a
3605b
3535ab
3473ac
3489ac
3532ab
3517ab
0-42
4637abc
4549ab
4721c
4582ab
4715c
4614abc
4526b
4573ab
4649ac
4645ac
0-21
796c
760bd
796 c
784ac
797c
767ab
743d
796c
776abc
781abc
22-42
1803d
1920a
1980 c
1894ab
1923a
1908ab
1907ab
1857bd
1898ab
1884ab
0-42
2595d
2681ab
2777c
2679ab
2720ac
2675ab
2651bd
2657bd
2674ab
2666ab
0-21
1.354d
1.430a
1.398bc
1.423ab
1.391c
1.418ab
1.432a
1.354d
1.443a
1.443a
22-42
1.986e
1.810bd
1.855ac
1.847abc
1.880a
1.854abc
1.831cd
1.882a
1.863ac
1.878a
0-42
1.788d
1.702b
1.719ab
1.721ab
1.735ac
1.728abc
1.718ab
1.722ab
1.739ac
1.749c
33.6
94.8
117.4
22.5
56.52
62.2
0.025
0.041
0.0261
Different superscripts in columns indicate significantly (P<0.05) different means; comparing wheat vs. sorghum.
1= Control wheat-based diet; 2= Sorghum H Pioneer Bonus, western downs Qld; 3= Sorghum M Pacific MR Buster (Liverpool plains, NSW), with reduced AME by 0.8MJ
in starter phase only to switch to sorghum M during finisher period using normal sorghum AME value; 4= Sorghum B (Pacific Buster, Lowood, Qld); 5= Sorghum E= Pacific
Buster (Lockyer, Qld) AME reduced by 0.8MJ starter phase only; 6= Commercial sorghum, purchased in Brisbane; 7= Sorghum K= Hylan Lyberty (Northern Downs, Qld);
8= Control wheat-based diet during starter period to switch to commercial sorghum during finisher period; 9= Commercial sorghum + phytase added on top; 10= Commercial
sorghum with phytase added after reducing available phosphorous (%) from 0.44 to 0.38 starter period and from 0.34 to 0.30 finisher period. Calcium was reduced (%) from
0.85 to 0.75 starter period and from 0.68 to 0.60 finisher period.
58
Aust. Poult. Sci. Symp. 2010...21
ENERGY UTILISATION AND DIGESTIBILITY OF FATS AS INFLUENCED BY THE
AGE OF BROILERS
P. TANCHAROENRAT 1, F. ZAEFARIAN1, G. RAVINDRAN1
and V. RAVINDRAN1
A study was conducted to investigate the influence of age of broilers on the apparent
metabolisable energy (AME) and total tract fat digestibility of different types of fats (tallow,
soybean oil, 50:50 mixture of tallow and soybean oil, poultry fat and palm oil). The assay
diets were developed by substituting the different fats for 4 % (w/w) of a maize-soybean meal
basal diet. The diets were offered ad libitum in mash form to six replicate cages of broilers (6
birds/cage) from day 1 to day 35 post-hatching. Total excreta collection was made during the
first, second, third and fifth weeks for the determination of AME. The influence of fat type
and age of birds on the AME (MJ/kg dry matter) and total tract digestibility coefficient of fats
is summarised in the table below.
Fat type
Tallow
Soybean oil
Tallow: soybean oil (50:50)
Poultry fat
Palm oil
SEM
Age of birds (week)
1
2
3
5
SEM
Probabilities, P ≤
Fat type
Age of birds
Fat type x age of birds
AME
Fat digestibility
24.4c
32.4ab
28.7bc
29.5ab
33.7 a
1.18
0.45c
0.73a
0.62b
0.64b
0.67b
0.015
17.4b
31.6a
35.5a
34.4a
1.06
0.39c
0.66b
0.71a
0.72a
0.013
***
***
NS
***
***
NS
NS, not significant; *** P < 0.0001.
a,b
Within each main effect, means in a column not sharing a common superscript are significantly
different (P < 0.05).
The results showed that these were no fat type x age interaction (P > 0.05), indicating
that the effect of age on the AME and total tract fat digestibility was similar for all fat types.
The AME of palm oil, soybean oil and poultry fat were determined to be high, whereas tallow
AME was lower (P < 0.05). Age of broilers significantly affected the AME of fats. The AME
was markedly lower (P < 0.05) during week 1, but improved during week 2. There were no
further improvements (P > 0.05) in the AME after week 2. The patterns observed for total
tract digestibility of fat were somewhat similar, highlighting the physiological limitation in
young birds to effectively digest and utilise fats and also confirm the poor digestibility of
tallow in poultry.
1
Institute of Food, Nutrition and Human Health, Massey University Palmerston North, New Zealand
59
Aust. Poult. Sci. Symp. 2010...21
TOOLS IN EARLY NUTRITION TO MAXIMISE GROWTH PERFORMANCE
A. KOCHER 1, A. NAYLOR1, C. MARTIN1, T. WILSON 2 and J. HAZELDENE 3
Summary
The effect of a commercial yeast extract was evaluated in a commercial broiler operation in
Victoria. Yeast extract was added to a commercial starter diet at 1.66% and growth
performance of birds was recorded over a 15 month period. The addition of yeast extract
resulted in small numerical improvements in weight gain and feed conversion ratio. An
economic analysis based on the actual performance data showed that despite the lack of
statistically significant differences in the evaluation, there was an economic benefit of 5 cents
per bird. In conclusion, this study showed that it is difficult to demonstrate statistically
significant differences in broiler performance when testing specific feed additives under
commercial conditions. However based on the findings in this study and comparable findings
from the literature the addition of yeast extract in broiler starter diets can provide a tool to
increase profitability for broiler producers.
I. INTRODUCTION
Genetic progress has enabled the broiler industry to grow broilers much faster and for birds to
achieve market weights at a much younger age. A key factor in achieving this potential is the
rapid and early development of the gastrointestinal tract, and in particular the development of
functional and mature villi. The actual formation of villi starts before hatch and the most
rapid growth of the intestine occurs in the last 3 days of incubation (Uni et al., 2003). Post
hatch the weight and size of the intestine increases more rapidly in comparison to other
organs. Within the first hours post hatch the crypts begin to form and intensive cryptogenesis
occurs. At the same time villi length increases rapidly before reaching a plateau around 8
days in the duodenum or 10 days in the jejunum and ileum (Sklan, 2004).
In addition to the physical growth of the intestine, enterocytes in the small intestine
mature and become fully functional. At hatch protein and lipids from yolk are the only
nutrients available to the chick; however pancreatic enzymes and brush border enzymes are
detected in the embryonic intestine before hatch (Uni et al., 2003) and pancreatic enzyme
activities, brush border enzyme activities and transporter activities continue increasing post
hatch (Sklan, 2004). Several authors reported a more rapid development of the intestine when
birds were given access to feed immediately post hatch (Ao and Choct, 2004; Sklan, 2003).
Parallel to the morphological development of the intestine, colonisation of the intestine with
microorganisms starts immediately after hatch. Comprehensive work by (Lee et al., 2006)
showed that the composition of the microbial community depends very much on the
digestibility of the diet and therefore the nutrients available to the bird and the microflora
respectively. Bacterial colonisation is also related to the response of the gut-associated
lymphoid system (GALT). The protection in the intestine is a combination of the innate
immune response and an adaptive immune response through the secretion of antibodies
(Sklan, 2004). Particularly in the early stages of development the GALT rapidly adapts to
distinguish antigens as part of the diet and antigens derived from pathogens.
Considering the requirement of nutrients for the development of the intestine and growth,
and the challenges in balancing the microflora and immune response, appropriate nutrition
1
Alltech Biotechnology Pty. Ltd, Dandenong South, VIC 3175 Australia
Scolexia, Moonee Ponds, VIC 3039 Australia
3
Hazeldene’s Chicken Farm, Lockwood, VIC 3551 Australia
2
60
Aust. Poult. Sci. Symp. 2010...21
and diet composition during the first week are crucial to maximise overall broiler
performance. It is proposed that using a pre-starter diet based on highly digestible nutrients
results in bodyweights 15-20% higher in comparison to a conventional starter diet (Leeson
and Summers, 2005). In addition to the supply of highly digestible nutrients it is important to
provide dietary nucleotides to support the rapid cell and tissue multiplication that occurs post
hatch. The chick has the ability to synthesise nucleotides, however under conditions of rapid
growth and increased immune challenge de novo synthesis appears to be inadequate (Rutz et
al., 2008). A potential source of dietary nucleotides for pre-starter diets is yeast extract. Using
a commercial source of yeast extract (NuPro), Zauk et al. (2006) found an improvement in
histological traits of villi in the intestine, however this did not result in any significant
improvements in broiler performance. A lack of environmental challenges is the most likely
explanation for this observation. This paper reports the result of a study investigating the
addition of yeast extract in the starter phase under commercial conditions in Australia.
II. MATERIALS AND METHODS
The current experiment was conducted on a commercial grow out farm (Hazeldene’s Chicken
Farm) near Bendigo, Victoria over a period of 15 months (August 2008 to October 2009).
Birds were raised in houses with identical layout and environmental conditions, and normal
bird management was applied across the entire farm. The control group received a
commercial four phase feeding program (starter, grower, finisher and withdrawal), with
change-over of feed at approx. 10 days, 26 days and 35 days of age. The experimental group
was fed a commercial starter diet with 1.66% yeast extract (NuPro®, Alltech Inc). All of the
other diets were the same as the control group.
The farm was divided into 10 commercial broiler grow out sheds with a capacity of
40,000 birds each. One feed silo was allocated to 2 sheds therefore each cycle comprised of 5
experimental units. A total of 6 cycles were included in the evaluation, as a result each
treatment was replicated 15 times over the 15 month period. Day old chicks (AA, Cobb and
Ross) were supplied from breeder flocks of different ages. Feed intake and weight gain was
measured at the end of the experiment and mortality was recorded on a weekly basis. Feed
conversion was standardised to 2.45kg bodyweight. At the end of the experiment an
economic analysis was conducted using the Alltech Poultry i-solution program (Francis and
Sacranie 2009). Data were analysed according to the GLM procedure for ANOVA. Duncan’s
multiple-range test was used to separate means when significant effects (P < 0.05) were
detected by analysis of variance.
III.
RESULTS
Overall performance in the current trial was comparable to industry standards. The addition
of yeast extract in the starter diet of broilers resulted in small numerical improvements in
broiler growth performance (Table 1). Average bodyweight at the end of the 42day grow–out
cycle was around 2.5kg in both groups, with numerical advantage of 35g (1.4%) for birds fed
yeast compared to the control. Similarly, the adjusted feed conversion ratio (corrected to
2.45kg life weight) was slightly improved (+3 points, 1.5%) for birds receiving a modified
starter diet, however these differences were not statistically significant.
An economic analysis based on the actual performance data clearly showed that
despite the lack of significant differences in the evaluation, there was a substantial economic
benefit for birds fed a yeast extract in the starter phase. Under current market conditions
(October 2009), the value per broiler sold based on feed cost and sales price was 5 cents
higher for birds fed yeast extract in the starter phase in comparison to birds fed a
61
Aust. Poult. Sci. Symp. 2010...21
conventional commercial starter diet. The return on extra outlay (REO) was calculated to be
2.4:1 when taking into account the cost of the additive.
Table 1.
Control
Yeast Extract
Growth performance of broilers on commercial diets with or without yeast
extract (NuPro) in the starter phase
Number of
birds
Final weight
1,160,015
1,154,561
2.544
2.579
FCR
FCR1
EPEF2
1.995
1.985
1.972
1.942
277
281
STD
0.120
0.075
0.097
Treatment
NS
NS
NS
1
FCR corrected to 2.45kg live weight, 2EPEF = European Poultry Efficiency Factor
Table 2.
21.2
NS
Economic analysis of broilers on commercial diets with or without yeast
extract (NuPro) in the starter phase
Number of birds
Cost of yeast extract per broiler2
Feed cost per broiler harvested
Margin over feed /broiler
Extra value per broiler sold
Control
Yeast Extract
1,160,015
$0.00
$2.03
$4.33
---
1,154,561
$0.02
$2.07
$4.38
$0.05
1
the price per kg live weight was set at $2.50 2yeast extract cost as supplied by the commercial
operator
IV. DISCUSSION and CONCLUSIONS
The benefits of highly digestible pre-starter diets has been the subject of a number of
scientific publications, however only a few authors looked specifically at the benefit of yeast
extract on lifetime performance of broilers (Leeson, 2008; Nollet et al., 2008; Rutz et al.,
2008; Santos et al., 2008; Zhang et al., 2005). Although it is generally recognised that the
addition of yeast extract can result in better growth performance all of these studies failed to
demonstrate a significant improvement in either weight gain or feed conversion ratio. For
example, Leeson (2008) reported that feeding a pre-starter diet with yeast extract, organic
minerals and Bio-Mos in the first 4 days resulted in a numerical improvement of 9% (2670g
vs 24560g) at 42days. Similarly, Nollet et al. (2008) showed that adding 2% yeast extract in
the first 2 weeks of age reduced the FCR by 3 points at 42days (1.632 vs 1.660). The findings
in the current experiment under commercial conditions are in line with these reports. The lack
of statistically significant differences in performance data is due to larger variation under
commercial conditions compared to trials conducted in university or research settings and the
limited number of replicates. Rutz et al. (2007) reported significant improvements in weight
gain and feed efficiency at 42 days by inclusion of 2% and 1.5% of yeast extract in pre-starter
diets under strictly controlled trial conditions.
62
Aust. Poult. Sci. Symp. 2010...21
Considering the number of studies in the literature which do show significant benefits
when using yeast extract an economic analysis was conducted to determine the actual cost
benefit of using yeast extract in the current experiment. Although the extract was only added
in the starter phase (day 0-10) no actual figures on exact feed intake in this period were
available, it is assumed that starter feed is 8% of the total feed consumed (Kenny and Kemp,
2003). Based on current market conditions (October, 2009) the benefit of adding yeast extract
to a starter diet is 5 cents per bird.
In conclusion, this study shows that it is difficult to demonstrate significant
differences in broiler performance when testing specific feed additives under commercial
conditions. Nevertheless, based on the findings in this study and comparable findings from
the literature the addition of yeast extract in broiler starter diets can provide a tool to increase
profitability for broiler producers.
REFERENCES
Ao Z, Choct M (2004) Proceedings, Australian Poultry Science Symposium 16, 116-119.
Francis M, Sacranie A (2009) In: Alltech Biotechnology. Stamford.
Kenny M, Kemp C (2003). In: International Poultry Production. pp. 12-14.
Lee MD, Lu J, Harmon B, Hofacre C, Maurer JJ (2006) In: Antimicrobial growth
promoters: Worldwide ban on the horizon?' (D Barug, J de Jong, AK Kies, MWA
Verstegen Eds) pp 149-163. Noordwijk aan Zee, The Netherlands.
Leeson S (2008) In: 'AllAboutFeed.net'.
Leeson S, Summers JD (2005) University Books. Guelph, Ontaria.
Nollet L, Geers R, Spring P (2008) World Poultry Congress. Brisbane, Qld. WPSA.
Rutz F, Concales X, Anciuti MA, Fernando V, Roll B, Rossi P (2007) Biotechnology in the
Feed Industry: Proceedings of Alltech's 23th Annual Symposium. (TP Lyons, KA
Jacques, JM Hower Eds) pp 175-181. Nottingham Press, Nottingham.
Rutz F, Concales X, Anciuti MA, Fernando V, Roll B, Rossi P (2008) In: Gut efficiency:
The key ingredient in pig and poultry production. (J Taylor-Pickard, P Spring, Eds)
pp 155-165. Wageningen Academic Press.
Santos AA, Jr., Bohorquez DV, Nanney RL, Ferket PR (2008) World Poultry Congress.
Brisbane, Qld. WPSA.
Sklan D (2003) Poultry Science 82, 117-122.
Sklan D (2004) In: Interfacing Immunity and Health. (L Tucker, JA Pickard-Taylor Eds).
Nottingham University Press.
Uni Z, Tako E, Gal-Garber O, Sklan D (2003) Poultry Science 82, 1747-1754.
Zauk NHF, Lopes DCM, Silva LM, Dallmann PR, Ribeiro CLG, Pinto AO, Mielke RB,
Anciuti MA, Rutz F (2006) Nutritional Biotechnology in the Food and Feed
Industry: Proceedings of Alltech's 22nd Annual Symposium. p 10. Lexington KY.
Zhang AW, Lee BD, Lee SK, Lee KW, An GH, Song KB, Lee CH (2005) Poultry Science
84, 1015-1021.
63
Aust. Poult. Sci. Symp. 2010...21
DIETS HIGH IN LINOLEIC ACID REDUCE OMEGA-3 LONG CHAIN
POLYUNSATURATED FATTY ACIDS IN CHICKEN TISSUES
L.R. KARTIKASARI 1,2, R.J. HUGHES 3, M.S. GEIER3, M. MAKRIDES 4 and R.A.
GIBSON1
Summary
We have previously demonstrated that the level of omega-3 long chain polyunsaturated fatty
acids (n-3 LCPUFA) could be increased several fold by increasing the level of alphalinolenic acid (ALA) in broiler feed. The lowest LA to ALA ratio of experimental diets
resulted in the highest n-3 LCPUFA, EPA, DPA and DHA in both breast and thigh tissues.
Because the n-6 PUFA, linoleic acid (LA), competes for the enzymes used to produce n-3
LCPUFA, the effect of dietary LA on n-3 LCPUFA accumulation in chicken meat was
unclear. The objective of this study was to examine the effect of varying LA levels in diets on
the conversion of ALA into EPA, DPA and DHA into chicken tissues. The level of ALA in
the diets was held constant at 2.1% energy (% en) while the level of LA varied from 2.9 to
4.4% en. The ratio of LA to ALA of the experimental diets thus ranged from 1.4:1 to 2.1:1.
The results indicated that the total n-3 LCPUFA levels in the breast meat of birds fed with the
lowest LA content was 16% higher than the n-3 LCPUFA in the breast of birds fed with the
highest LA content. In general, the decrease in n-3 LCPUFA due to inhibition by LA was less
than the stimulatory effect of an equivalent level of ALA on n-3 LCPUFA accumulation.
This study indicated that the strongest influence on n-3 LCPUFA accumulation in chicken
tissues was the level of ALA in the diet. The experimental diets did not appear to affect the
growth performance of chickens. We conclude that there was only a modest effect of dietary
LA on omega-3 LCPUFA accumulations in chicken meat, but diets that are lower in LA will
allow greater conversion of ALA into n-3 LCPUFA.
I. INTRODUCTION
There have been a number of reports of n-3 fatty acid enrichment of chicken meat through
dietary enrichment with vegetable oils containing the n-3 precursor, ALA; however, the
levels of n-3 LCPUFA found in meat have been low (Febel et al., 2008; Zelenka et al., 2008).
Recently we have demonstrated that diets with high ALA content increased the incorporation
of EPA and DHA into breast and thigh meat to levels 5 and 4-fold relative to birds fed low
ALA (Kartikasari et al., 2009). The levels of n-3 LCPUFA achieved in our study were much
higher than levels reported previously (Febel et al., 2008; Zelenka et al., 2008). The
improvement in meat n-3 LCPUFA status was achieved by increasing the level of ALA in the
diets and keeping a constant LA level in order to optimize the conversion of ALA into EPA
and DHA. However, LA and ALA compete as substrate for desaturation and elongation
enzymes in fatty acid metabolism (Liou et al., 2007). The results of this study indicated that
the conversion of ALA was mainly driven by the level of ALA in the diet. Experimental diets
with the lowest LA to ALA ratio resulted in the highest EPA, DPA and DHA in these tissues.
1
School of Agriculture, Food and Wine, University of Adelaide, SA
Department of Animal Product Technology, Faculty of Agriculture, Sebelas Maret University,
Indonesia
3
Pig and Poultry Production Institute, South Australian Research and Development Institute,
Roseworthy Campus, SA
4
School of Paediatrics and Reproductive Health, University of Adelaide, SA.
2
64
Aust. Poult. Sci. Symp. 2010...21
These results indicated that increasing the ALA content of the diets was important but
it was unclear whether keeping a constant LA level is also important in the regulation of
EPA, DPA and DHA accumulation in chicken meat. The objective of the current study was to
examine the effect of LA levels in diets on the conversion of ALA into EPA, DPA and DHA
into chicken tissues.
II. MATERIALS AND METHODS
One-day-old mixed sex broiler chickens (48 Cobb 500) were used in this study. The birds
were randomly housed in six pens (n=8 birds per pen) and distributed among three dietary
treatments (2 pens for each dietary treatment) and reared for 28 days under controlled
environmental conditions. The experiment was approved by the Animal Ethics Committees
of the Department of Primary Industries and Resources South Australia and the University of
Adelaide.
The three diets were based on a standard commercial starter diet, with a low level of
fat. The dietary treatments had the same nutritional values as the basal diet except for the
fatty acid composition of the fats. Pure or blended vegetable oils including macadamia,
flaxseed and sunflower oils were included at a level of 2.8% in order to produce diets with
the desired levels of LA and ALA. The LA level of control diet was 2.3% en with low ALA
level (0.2% en). The experimental diets were formulated by varying the levels of LA, which
ranged from 2.9 to 4.4% en, with ALA levels kept constant at 2.1% en. The total fat content
was kept constant at approximately 5%. Each diet was provided ad libitum for the duration of
the 28-d growth period. At 28 days post-hatch, six selected birds from each pen (12 birds per
group) were weighed individually and breast and thigh tissues were collected. Total lipids in
diet and tissue samples were extracted following the method of Folch et al. (1957) and the
phospholipids were methylated as described by Blank et al. (2002). The fatty acid
composition of tissue samples was determined and quantified by gas chromatography.
III. RESULTS
Elevating the level of dietary LA had an effect that was not quite significant on ALA levels in
breast tissues and only reached significance in thigh tissue (P < 0.05). The diets caused a
consistent reduction in EPA levels in all tissues (P < 0.01; Table 1 and 2), but DPA and DHA
levels were unaffected by dietary LA. While increased levels of LA tended to decrease the
level of total n-3 PUFA (P < 0.1), the level of total n-6 fatty acids increased (P < 0.01). The
highest level of dietary LA (4.4% en) resulted in the highest level of total breast and thigh
phospholipids n-6 fatty acids, which was 27.28 and 29.15% of total fatty acids, respectively.
Elevating dietary LA increased the level of LA in all tissues but this was not translated to
changes in AA levels. The increased levels of dietary LA from 2.9 to 4.4% en did not affect
tissue fat content including breast and thigh and final weight of birds at 28 days of age.
IV. DISCUSSION
The results observed in this study indicated that increasing the dietary concentration of LA as
energy from 2.9 to 4.4% whilst keeping a constant ALA level reduced tissue (breast and
thigh) n-3 LCPUFA (EPA) accumulation; however DPA and DHA tended to not be affected.
The decrease in n-3 LCPUFA levels might be due to competition between ALA, the
precursor to n-3 LCPUFA, and LA for Δ6-desaturase. Thus, a high dietary level of LA might
depress the conversion of ALA to n-3 LCPUFA (Arbuckle et al., 1992; Liou et al., 2007).
65
Aust. Poult. Sci. Symp. 2010...21
Table 1.
Fatty acid composition of breast phospholipids from chickens fed
experimental diets varying in LA while holding ALA constant (2.1% en) for
28 days
Experimental diets
Control ( 0.2% en 2.9 % en LA
ALA)
Fatty acids (%)
SFA
MUFA
18:2n-6
20:4n-6
Total n-6
18:3n-3
20:5n-3
22:5n-3
22:6n-3
Total n-3
Total PUFA
35.63
38.08
14.33
3.48
21.39
0.40
0.62
0.85
0.94
3.09
24.48
38.36a
26.51b
14.80a
3.62
21.63a
1.52b
2.63b
4.24
3.23
12.66
34.29b
4.4% en LA
39.40b
21.56a
18.31b
4.78
27.28b
1.34b
2.00a
4.11
2.63
11.09
38.37c
PSEM
0.107
0.825
0.251
0.255
0.395
0.061
0.054
0.268
0.108
0.366
0.750
P value
*
*
*
NS
*
NS
*
NS
NS
NS
NS
Values are means of twelve observations (n = 12) per treatment and their pooled standard error of the
mean (PSEM). a,bValues in the same row with no common superscript are significantly different (P <
0.05). SFA = saturated fatty acid; MUFA = monounsaturated fatty acid; PUFA = polyunsaturated
fatty acid. NS =not significant; *P < 0.05.
Table 2.
Fatty acid composition of thigh phospholipids from chickens fed experimental
diets varying in LA while holding ALA constant (2.1% en) for 28 days
Experimental diets
Control (0.2% en 2.9% en LA 4.4% en LA
ALA)
Fatty acids (%)
SFA
MUFA
18:2n-6
20:4n-6
Total n-6
18:3n-3
20:5n-3
22:5n-3
22:6n-3
Total n-3
Total PUFA
34.81
39.10
14.22
4.49
21.67
0.32
0.52
0.98
0.83
2.85
24.53
37.40a
27.32b
16.67a
4.27
23.49a
1.48b
2.15b
4.12
2.63
11.10
34.60a
38.52b
22.27a
20.42b
5.37
29.15b
1.25a
1.46a
3.93
2.11
9.53
38.69b
PSEM
0.089
0.692
0.354
0.214
0.361
0.023
0.059
0.172
0.100
0.310
0.670
P value
*
*
*
NS
**
*
*
NS
NS
NS
*
Values are means of twelve observations (n = 12) per treatment and their pooled standard error of the
mean (PSEM). a,bValues in the same row with no common superscript are significantly different (P <
0.05); *P < 0.05, **P < 0.01.
66
Aust. Poult. Sci. Symp. 2010...21
Although increasing the proportion of LA in the diet appeared to decrease n-3
LCPUFA accumulation in tissue samples, the effect was similar in size to the increases in n-3
LCPUFA caused by increasing dietary ALA by a similar amount. The total n-3 LCPUFA
levels in breast and thigh of birds fed with the lowest LA content (2.9% en) was 15.6 and
18.7%, respectively higher than the n-3 LCPUFA in the breast and thigh of birds fed the
highest LA content (4.4% en). These findings indicate that the effect of LA content in the
diets is important in regulating the n-3 LCPUFA accumulation in tissues. The changes found
were in concordance with other authors (Marangoni et al., 1992; Liou et al., 2007; Zelenka et
al., 2008). Zelenka et al. (2008) reported that there was no effect on the level of DPA and
DHA both in breast and thigh meat by increasing dietary levels of LA in the diet of chickens
from 14.2 to 58.4% of total fatty acids. Conversely, we found that with increasing LA content
in the diet, the level of LA rose in tissue samples, but n-6 LCPUFA, AA, did not respond.
These findings also agreed with studies reported previously (Marangoni et al., 1992; Hussein
et al., 2005).
The increased levels of dietary LA did not affect tissue fat content. This result is in
concordance with those of Zelenka et al. (2008) using increased levels of dietary LA. Their
results show that there was no difference on fat content of breast by increased levels of
dietary LA from 14.22 to 58.44 g/kg. The experimental diets did not influence significantly
the final weight of birds. A number of investigators (Febel et al., 2008; Kavouridou et al.,
2008) have reported similar findings regarding the final weight among broilers fed different
types of fat sources involving vegetable oils.
In summary, increasing the levels of dietary LA reduced tissue n-3 LCPUFA (EPA
and to a lesser extent DPA and DHA) concentration. In addition, the concentration of LA and
total n-6 fatty acids appeared to be strongly influenced by the level of dietary LA while the
accumulation of AA tissue samples did not change. We conclude that diets that are lower in
LA will allow greater conversion of ALA into n-3 LCPUFA.
REFERENCES
Arbuckle LD, Rioux FM, MacKinon MJ, Innis SM (1992) Biochimica Et Biophysica Acta
1125, 262-267.
Blank C, Neumann MA, Makrides M, Gibson RA (2002) Journal of Lipid Research 43,
1537-1543.
Febel H, Mezes M, Palfy T, Herman A, Gundel J, Lugasi A, Balogh K, Kocsis I, Blazovics
A (2008) Journal of Animal Physiology Animal Nutrition 92, 369-376.
Folch J, Lees M, Sloane Stanley GH (1957) Journal of Biological Chemistry 226, 497-509.
Hussein N, Ah-Sing E, Wilkinson P, Leach C, Griffin BA, Millward DJ (2005) Journal of
Lipid Research 46, 269-280.
Kartikasari LR, Hughes RJ, Geier MS, Makrides M, Gibson RA (2009) World Congress on
Oils and Fats 28th ISF Congress. Pp 62-63.
Kavouridou K, Barroeta AC, Villaverde C (2008) Spanish Journal of Agricultural Research
6, 210-218.
Liou YA, King DJ, Zibrik D, Innis SM (2007) Journal of Nutrition 137, 945-952.
Marangoni F, Mosconi C, Galella G, Galli C (1992) Lipids 27, 624-628.
Zelenka J, Schneiderova D, Mrkvicova E, Dolezal P (2008) Veterinary Medicine 53, 77-85.
67
Aust. Poult. Sci. Symp. 2010...21
THE VULNERABILITY OF SORGHUM TO ‘MOIST-HEAT’
P.H. SELLE 1, R.J. GILL1 and J.A. DOWNING1
Summary
A hydrothermal procedure involving microbial phytase and citric acid reduced the phytate
content of sorghum by 89.8%. A slurry of equal parts finely-ground sorghum and distilled
water was mixed for 2 hours at 45°C followed by drying at 60°C for 70 hours. However, both
dephytinised and sham-treated sorghum significantly depressed growth performance and N
retention to very similar extents. Collectively, the hydrothermal procedure reduced weight
gain by 27.0%, feed intake by 20.3% and feed efficiency by 9.5% in broilers from 7-21 days
post-hatch and N retention was reduced by 8.0% or 5.05 percentage units. The potential
benefits of dephytinisation were presumably overwhelmed by the vulnerability of the
nutritive value of sorghum to moist-heat as demonstrated in broiler chickens in this study, the
implications of which are considered.
I. INTRODUCTION
A hydrothermal procedure was developed to dephytinise sorghum with microbial phytase and
citric acid. An important prerequisite was that placebo or sham-treatment needed to be
innocuous but an evaluation of sham-treated sorghum in broilers demonstrated that this
exposure to moist-heat significantly depressed growth performance and nitrogen (N)
retention. The unique vulnerability of the in vitro protein digestibility of sorghum to wetcooking procedures, or ‘moist-heat’, has been extensively documented; however, few
assessments of the impact of wet-cooked sorghum on growth performance and nutrient
utilisation in broilers have been reported.
II. MATERIALS AND METHODS
The hydrothermal dephytinisation procedure involved mixing finely hammer-milled sorghum
(1 mm sieve) with an equal quantity of distilled water plus 10,000 FTU/kg Aspergillus niger
phytase and 3 g/kg citric acid. The slurry was constantly agitated for 2 hours at 45°C and then
oven-dried at 60°C for 70 hours to a moisture content of approximately 5%. Sham-treatment
followed the same procedure but without the addition of phytase and citric acid. Citric acid
reduced the pH of the slurry from 6.24 to 4.89. Phytate concentrations were determined by a
‘ferric chloride-precipitation’ method described by Miller et al. (1980). The procedure
reduced the phytate content of sorghum by 89.8%, from 10.04 to 1.03 g/kg on a dry matter
basis.
Three sorghum-casein diets containing either control, sham-treated or dephytinised
sorghum were prepared in which citric acid was either added as a dietary ingredient or
incorporated into the dephytinised sorghum. Each of the dietary treatments was offered to 7
replicates (6 birds per cage) of male Cobb chicks from 7-21 days post-hatch to determine
treatment effects on growth performance. Total excreta were collected over the final 72 hours
to determine apparent metabolisable energy (AME) and N retention by standard procedures.
Experimental data was analysed by one-way analysis of variance using general linear
model procedures (SPSS Inc.). Least significant differences were calculated to separate
1
Faculty of Veterinary Science, The University of Sydney. Camden NSW 2570
68
Aust. Poult. Sci. Symp. 2010...21
mean values where the effect of treatment was significant at the 5% level of probability. The
conduct of the experiment complied with specific guidelines set down by the Animal Ethics
Committee of Sydney University.
III. RESULTS
As shown in Table 1, both dephytinised and sham-treated sorghum significantly depressed
growth performance (P < 0.001) and N retention (P < 0.02) to very similar extents in
comparison to untreated sorghum. Collectively, exposure of sorghum to moist-heat reduced
weight gain by 27.0% (357.5 versus 490 g/bird), feed intake by 20.3% (659.5 versus 827
g/bird) and feed efficiency by 9.5% (1.85 versus 1.69) from 7-21 days post-hatch. Also, N
retention was reduced by 5.05 percentage units or 8.0% (57.675 versus 62.72%), while AME
was numerically depressed.
Table 1.
Treatment
Sorghum
Untreated
Sham-treated
Dephytinised
Effects of moist-heated (sham-treated and dephytinised) sorghum on growth
performance and nutrient utilisation of broilers.
Growth performance
Weight
Feed
Feed
gain
intake
efficiency
(g/bird)
(g/bird)
(g/g)
490a
352b
363b
827a
650b
669b
1.69a
1.86b
1.84b
Nutrient utilisation
AME
N
(MJ/kg
retention
DM)
(%)
15.48
15.28
15.33
62.72a
57.36b
57.99b
0.0689
SEM
15.75
21.38
0.0262
1.246
0.128
Significance (P =)
0.000
0.000
0.000
0.013
LSD (P < 0.05)
46.8
63.5
0.077
3.701
ab
Means with different superscripts within columns are significantly different (P < 0.05)
IV. DISCUSSION
Clearly, sham-treatment of sorghum under the conditions described was not
innocuous and the hydro-thermal process is not suitable for the evaluation of dephytinised
sorghum to define the anti-nutritive effects of phytate. In this context, the caution issued by
Harland and Oberleas (1987) is instructive because they concluded that there is no practical
way to remove phytate from feed ingredients without altering nutrient bioavailability.
However, the addition of citric acid to the slurry facilitates the enzymic degradation of
phytate (Selle, 2006), which is probably mainly achieved via the reduction in slurry pH from
6.24 to 4.89 pH. Phytate is mainly present in feedstuffs as a mineral-phytate complex
involving magnesium (Mg) and potassium (K) as Mg5-K2-phytate (Lott et al., 2004). The pH
reduction would substantially increase the solubility of magnesium-phytate complexes in
sorghum (Cheryan et al., 1983) and facilitate the enzymic degradation of phytate.
Nevertheless, the potential benefits of dephytinisation were presumably overwhelmed by the
vulnerability of the nutritive value of sorghum to moist-heat under the conditions of this
experiment.
Sorghum is perhaps uniquely vulnerable to moist-heat as Hamaker et al. (1987)
reported that wet-cooking reduced the in vitro pepsin digestibility of sorghum by 24.5%
(0.563 versus 0.808) in contrast to more modest reductions in maize (4.1%), wheat (5.4%),
69
Aust. Poult. Sci. Symp. 2010...21
rice (9.1%) and barley (13.0%). As reviewed by Duodu et al. (2003), exposure of sorghum to
moist-heat induces disulphide cross-linkages and conformational changes particularly in βand γ-kafirin and this, in turn, reduces the digestibility of α-kafirin which is centrally located
in protein bodies in the endosperm. Recently, Emmambux and Taylor (2009) reported that
boiling a 5:1 mixture of distilled water and sorghum for 30 minutes reduced in vitro pepsin
digestibility of kafirin by 24% (0.576 versus 0.760); however, pepsin digestibility of sorghum
protein was similarly reduced by 27% (0.599 versus 0.826). Interestingly, these results
suggest that the ‘non-kafirin’ component of sorghum protein, which is predominantly
glutelin, is equally vulnerable to moist-heat.
Few in vivo assessments of wet-cooked sorghum have been reported. However, moistheating a tannin-free sorghum depressed protein digestibility in rat balance studies and
reduced total tract digestibility of lysine and threonine by approximately 10% (Knudsen et
al., 1988). In the Mitaru et al. (1985) study, moist-heat consisted of boiling a 3:1 slurry of
distilled water and a low-tannin sorghum for 50 minutes and the impact of moist-heat on true
ileal digestibility (TID) of sorghum amino acids in broilers was determined. This procedure
reduced the average TID coefficients of 15 amino acids by 30.7% (0.629 versus 0.908) in
sorghum.
Wet-cooking did not significantly influence the AME of sorghum in the present
experiment, which suggests that starch digestibility was not altered. This is somewhat
surprising as Ezeogu et al. (2008) concluded that disulphide cross-linkages in the protein
matrix induced by wet-cooking limits expansion of starch granules and access of amylase to
its substrate and compromises starch digestibility.
Steam-pelleting broiler diets at high temperatures may have detrimental effects on
broiler performance (Raastad and Skrede, 2003; Creswell and Bedford, 2006). Moreover, the
starch gelatinisation temperature for sorghum is higher than maize and wheat (Taylor and
Dewar, 2001), which suggests that sorghum-based broiler diets are steam-pelleted at higher
temperatures. Thus, there is the possibility that steam-pelleting sorghum-based broiler diets at
high temperatures may constitute sufficient moist-heat to compromise the nutritive value of
sorghum. However, as discussed by Taylor (2005), relevant investigations do not appear to
have been completed.
It is noteworthy, however, that Zhuge et al. (1990) reported that increasing wetextrusion temperatures of sorghum from 105 to 132.5ºC, prior to its dietary incorporation,
significantly depressed weight gain by 17.0% (1235 versus 1488 g/bird) and feed efficiency
by 11.7% (1.883 versus 1.686) in a 35-day broiler bioassay. The performance of birds offered
sorghum extruded at 105ºC was comparable to the other processing methods (hammermilling, roller-milling, flaking) evaluated. This could indicate that steam-pelleting sorghumbased diets at 90-95ºC may not be deleterious; nevertheless, the Zhuge et al. (1990) study
demonstrates that exposing sorghum to sufficient moist-heat has the capacity to depress
broiler growth performance. This was shown in the present experiment with the indication
that protein utilisation was compromised, probably by the induction of disulphide crosslinkages, particularly in the kafirin protein fraction of sorghum.
In conclusion, an evaluation of the impact of steam-pelleting temperatures on growth
performance and nutrient utilisation of broilers offered sorghum-based diets appears justified.
Additional factors to be taken into consideration should include grain texture, grinding
method and particle size, and pellet ‘hardness’. Parsons et al. (2006) reported that broilers
offered a hard pellet (1856 g pellet breaking force) performed significantly better than
broilers on soft pellets (1662 g pellet breaking force). One possibility is that broilers may
‘prefer’ hard pellets (Nir, 1987) and the second is that hard pellets may stimulate gizzard
function in an analogous manner to whole grains (Cumming, 1994).
70
Aust. Poult. Sci. Symp. 2010...21
V. ACKNOWLEDGEMENT
The funding provided by the Chicken-meat Committee of the Rural Industries Research and
Development Corporation (RIRDC) is gratefully acknowledged.
REFERENCES
Cheryan M, Anderson FW, Grynspan F (1983) Cereal Chemistry 60, 235-237.
Creswell D, Bedford M (2006) Proceedings, Australian Poultry Science Symposium 18, 1-6.
Cumming RB (1994) Proceedings, 9th European Poultry Conference II, 219-222.
Duodu KG, Taylor JRN, Belton PS, Haymaker BR (2003) Journal of Cereal Science 38, 117131.
Emmambux NM, Taylor JRN (2009) Journal of the Science of Food and Agriculture 57,
1045-1050.
Ezeogu LI, Duodu KG, Emmambux MN, Taylor JRN (2008) Cereal Chemistry 85, 397-402.
Hamaker BR, Kirleis AW, Butler LG, Axtell JD, Mertz ET (1987). Proceedings of the
National Academy of Sciences of the United States of America 84, 626-628.
Harland BF, Oberleas D (1987) World Review of Nutrition and Dietetics 52, 235-239.
Knudsen KEB, Munck L, Eggum BO (1998). British Journal of Nutrition 59, 31-47.
Lott JNA, Ockenden I, Raboy V, Batten GD (2000) Seed Science Research 10, 11-33.
Miller GA, Youngs VL, Oplinger ES (1980) Cereal Chemistry 57, 189-191.
Mitaru BN, Reichert RD, Blair R (1985) Poultry Science 64, 101-106.
Nir I (1987) Proceedings, 6th European Symposium on Poultry Nutrition pp B21-B25.
Parsons AS, Buchanan NP, Blemings KP, Wilson ME, Moritz JS (2006) Journal of Applied
Poultry Research 15, 245-255.
Raastad N, Skrede A (2003) Proceedings, 14th European Symposium on Poultry Nutrition
115-116.
Selle PH (2006) Proceedings, Australian Poultry Science Symposium 18, 91.
Taylor JRN, Dewar J (2001) Advances in Food and Nutrition Research 43, 217-264.
Taylor JRN (2005) Proceedings, Australian Poultry Science Symposium 17, 9-16.
Zhuge Q, Yan Z, Klopfenstein CF, Behnke KC (1990) Feedstuffs 62, (9) 18, 55.
71
Aust. Poult. Sci. Symp. 2010...21
THE NUTRITIVE VALUE OF HIGH-YIELDING TRITICALE VARIETIES AND THEIR
POTENTIAL FOR INCLUSION IN POULTRY DIETS
A. WIDODO 1, P. IJI1 and J.V. NOLAN1
Triticale is a cereal grain that holds great promise as an alternative to wheat and other conventional
grains used in poultry diets. Triticale generally has a higher yield than wheat and adapts to more
difficult agronomic conditions than wheat (Korver et al., 2004). A crop breeding group at the
University of New England (UNE) has developed varieties that are even more high-yielding and
more disease-resistant than the current commercial strains. These varieties will need further
evaluation to establish their potential for animal, and particularly poultry feeding.
Eight varieties of triticale were obtained from the breeding group and subjected to detailed
and proximate analysis, prior to feeding trials. Cultivar H116 contained more protein than the other
cultivars (139 g/kg) while the lowest protein content was observed in cultivar H127 (Table 1). The
cultivars were very similar in lysine content, containing between 4.3 g/kg in cultivar H127 and 4.9
g/kg in H249. Methionine content varied from 1.6 (cultivar H127) to 2.0 g/kg in cultivar H249. The
gross energy content of the grains ranged from 18.2 (H127 and 128) to 18.5 MJ/kg (H55); the other
five cultivars being iso-caloric at 18.4 MJ/kg. Crude fat was also highest (27.4 g/kg) in cultivar
H55 and lowest in H128, and this may be the major cause of differences in energy values of
between these two cultivars. Total starch varied from 578 g/kg in cultivar H157 to 657 g/kg in
H249. Cultivar H55 was the highest in non-starch polysaccharides, 139 g/kg while H20 contained
only 90 g/kg.
Calcium content varied from 0.3 to 0.5 g/kg, respectively in cultivars H426 and H55, while
the phosphorus content was highest (4.4 g/kg) in cultivar H157 and lowest (3.5 g/kg) in H20.
Phytate was quite high in all cultivars, generally close to half of the total P content. Cultivar H20
had the lowest level of phytate, 1.8 g/kg, while the highest amount, 2.2 g/kg, was found in H249.
Table 2 shows the digestibility of dry matter, starch and viscosity of samples during in vitro
digestion. The in vitro method was adapted from Babynsky et al (1990), with slight modifications.
The in vitro dry matter digestibility (IVDMD) varied between 71.1 % (H55) and 77.5 % (H128).
Starch digestibility was between 19.5 % (H249) and 40.3 % (H157), and this low in vitro
digestibility may be due to the high content of resistant starch. Cultivar H55 was the most viscous
during digestion (1.2 cP) and this may be due to the high concentration of NSP in this cultivar.
REFERENCES
Babinsky L, Van der Meer JM, Boer H, Hartog (1990) J Sci Food Agric. 50, 173-178
Korver DR, Zuidhof MJ Lawes KR (2004) Poult. Sci. 83, 717-725.
1
School of Environmental and Rural Science, University of New England, Armidale NSW 2351.
72
Aust. Poult. Sci. Symp. 2010...21
Tabel 1.
Chemical composition (g/kg, dry matter) of the different varieties of triticale
Component
Dry matter
Ash
Crude protein
Gross Energy (MJ/kg)
Crude fat
Ca
P
K
Mn (mg/kg)
Na (mg/kg)
Total Starch
Resistant starch
P-Phytate
Total NSP
Soluble NSP
Insoluble NSP
Free sugars
Amino Acids
Arginine
Threonine
Alanine
Lysine
Methionine
Valine
Isoleucine
Leucine
Phenylalanine
Tabel 2.
H116
879
19.2
139
18.4
18.5
0.4
4.1
5.7
45.7
26.5
612
442
2.0
102
10
92
31
H127
873
17.4
125
18.2
15.9
0.4
4.0
5.5
59.6
16.6
654
478
2.1
104
12
91
27
7.5
4.4
5.9
4.6
1.9
6.5
4.9
9.1
6.2
6.6
3.9
5.1
4.3
1.6
5.5
4.1
8.0
5.6
H128
873
18.1
135
18.2
13.9
0.4
4.1
5.8
44.1
33.1
639
484
2.1
100
10
89
29
Variety
H157
873
18.7
136
18.4
14.6
0.4
4.4
5.7
59.7
35.0
578
463
2.2
119
10
110
23
7.6
4.6
5.7
4.8
1.8
6.5
4.8
9.3
6.6
7.4
4.4
5.5
4.6
1.9
6.4
4.8
9.2
6.4
H20
872
18.1
135
18.4
16.3
0.4
3.5
5.9
53.5
17.7
612
455
1.8
90
11
79
31
H249
875
17.5
136
18.4
17.5
0.4
3.6
5.4
49.4
38.8
657
509
2.2
96
9
87
27
H426
871
17.5
132
18.4
19.1
0.3
4.0
5.7
45.9
13.8
626
453
2.0
100
10
90
33
H55
873
16.7
126
18.5
27.4
0.5
3.7
5.0
47.3
30.1
592
453
1.9
139
11
128
28
7.3
4.3
5.8
4.7
1.9
6.4
4.7
8.9
6.0
7.9
4.7
6.0
4.9
2.0
6.8
5.0
9.5
6.5
7.6
4.4
5.6
4.7
2.0
6.4
4.8
9.0
6.4
7.4
4.5
5.4
4.6
1.8
6.1
4.5
8.8
6.1
In vitro digestibility of samples
Component
H116
H127
H128
Variety
H157
H20
H249
H426
H55
Dry matter digestibility (%)
75.8
76.3
77.5
74.0
75.4
77.3
77.4
71.1
Starch digestibility (%)
22.0
24.4
25.8
40.3
20.9
19.5
39.4
30.2
1.0
1.1
1.0
1.0
1.0
1.1
1.0
1.2
Viscosity during digestion (cP)
73
Aust. Poult. Sci. Symp. 2010...21
INFLUENCE OF HOUSING SYSTEMS ON THE BACTERIOLOGICAL QUALITY AND
SAFETY OF TABLE EGGS
K. DE REU 1*, W. MESSENS1, K. GRIJSPEERDT1, M. HEYNDRICKX1,
B. RODENBURG 2, M. UYTTENDAELE 3 and L. HERMAN1
Summary
With the introduction of alternative housing systems for laying hens in the EU, recently more
research focused on the bacterial contamination of table eggs, e.g. eggshell and egg content
contamination. Contamination of eggshells with aerobic bacteria is generally higher for nest
eggs derived from non-cage systems compared to furnished cages or conventional cages.
Studies indicate limited or no systematic differences in eggshell contamination with aerobic
bacteria between eggs laid in the nest boxes of furnished cages and eggs laid in conventional
cages. The major differences found in experimental studies between cage- and non-cage
systems are less pronounced under commercial conditions. The effect of housing system on
eggshell contamination with specific groups of bacteria is variable. Limited information is
available on the influence of housing system on egg content contamination. Recent research
does not indicate large differences in egg content contamination between eggs from cage- and
non-cage systems (ignoring outside nest and floor eggs). The microflora of the eggshell is
dominated by Gram-positive bacteria, whereas Gram-negative bacteria are best equipped to
overcome the antimicrobial defences of the egg content. Much of the research on eggshell
and egg content contamination focuses on Salmonella, since infection with Salmonella
Enteritidis, resulting from the consumption of contaminated eggs or egg products, is still a
major health problem. Observed Salmonella prevalence on the eggshell and in the egg
content vary, depending on the fact whether investigations were based on randomly sampled
table eggs or on eggs from naturally infected hens. Studies also show that it is highly unlikely
that a move from conventional cages to alternative cage systems and non-cage housing
systems will result in an increase in Salmonella infection and shedding, rather the opposite is
expected.
I. INTRODUCTION
The shell can become contaminated when passing through the vent, but many researchers
suggest that the main bacterial contamination occurs within a short period after laying due to
contact with dirty surfaces (Quarles et al., 1970; Gentry and Quarles, 1972). Messens et al.
(2005) and De Reu et al. (2006a; 2006b, 2006d) reported that increasing numbers of microorganisms on the eggshell consequently increase the risk of microbial eggshell penetration
and egg content contamination. Beside this horizontal route of bacterial infection of eggs, egg
contamination also occurs through the vertical or transovarian route. In the transovarian route
(vertical transmission), the yolk membrane (very infrequently the yolk itself), the albumen
1
Institute for Agricultural and Fisheries Research (ILVO), Technology and Food Science Unit,
Brusselsesteenweg 370, 9090 Melle, Belgium
2
Animal Breeding and Genomics Centre, Wageningen University, P.O. Box 338, 6700 A
Wageningen, The Netherlands
3
Department of Food Safety and Food Quality, Laboratory of Food Microbiology and Food
Preservation, University of Ghent, Coupure Links 653, 9000 Ghent, Belgium.
74
Aust. Poult. Sci. Symp. 2010...21
and/or the shell membranes are directly contaminated as a result of bacterial infection of the
reproductive organs.
II. GLOBAL BACTERIAL EGGSHELL AND EGG CONTAMINATION
a) Effect of the housing system
In early studies, bacterial eggshell contamination has been compared in litter and wire floor
houses. Quarles et al. (1970) reported that litter floor houses had on average nine times more
bacteria in the air, and 20 to 30 times more aerobic bacteria on the shell than wire floor
houses. Harry (1963) reported that the shells of eggs from deep litter systems had 15 times
more bacteria and a higher proportion of potential spoilage organisms than eggs from battery
cage systems.
Conventional cage housing for laying hens will be prohibited from 2012 in the
European Union, following EU-directive 1999/74. From 2012 onwards, only furnished cages
and non-cage systems (aviaries and floor housing) will be allowed. This has driven the recent
attention towards the effect of housing system on the bacterial eggshell contamination of
table eggs.
b) Conventional and furnished cages
De Reu et al. (2005b) compared the bacterial eggshell contamination of eggs laid in
conventional cages with eggs laid in the nest boxes of furnished cages. No systematic
difference in shell contamination with total counts of aerobic bacteria was found between
these systems (ranging from 4.0 – 4.5 log CFU/eggshell). Also, for Gram-negative bacteria
no difference was detected (both means ca. 3.0 log CFU/eggshell). The type of nest-floor
material in the nest boxes of the furnished cages also did not systematically influence the
bacterial eggshell contamination. Cepero et al. (2000; 2001) also found no differences in
counts of aerobic mesophilic bacteria but reported a higher prevalence of coliforms on shells
of eggs laid in furnished cages. Mallet et al. (2006) studied the hygienic aspects of eggs laid
at different locations in furnished cages. A significant difference in total count of aerobic
bacteria was observed on the eggshell of eggs collected from furnished cages (4.83 log
CFU/eggshell) compared to conventional cages (4.56 log CFU/eggshell). This was mainly
due to the eggs laid outside the nest in the litter area (4.96 log CFU/eggshell) or in the cage
(4.94 log CFU/eggshell). The bacterial load on eggs laid in the nests was similar to those
collected from the conventional cages. Similar conclusions were obtained for Enterococcus.
Wall et al. (2008) also found a higher bacterial load on eggs from furnished cages compared
to conventional cages. The bacterial counts were significantly (P < 0.001) higher in the
furnished cages compared to the conventional cages as regards Enterococcus and total
number of aerobic bacteria.
c) Cage- and non-cage systems
In further experimental studies, it was found that eggs from aviaries were contaminated with
higher numbers of aerobic bacteria than eggs from cage systems (Protais et al., 2003a; De
Reu et al., 2005b). The difference was more than 1 log unit (up to 5.1 – 6.0 log CFU/eggshell
for eggs from aviaries), with much higher counts on those eggs laid on the floor of the
aviaries (up to 7 log CFU/eggshell). For Gram-negative bacteria no systematic differences
were found between cage and non-cage housing systems (De Reu et al., 2005b).
d) Experimental studies compared to on farm studies
De Reu et al. (2005a; 2006c) evaluated whether the differences in initial eggshell
contamination, found in the experimental housing systems, were also applicable to
75
Aust. Poult. Sci. Symp. 2010...21
commercial conventional cage and non-cage housing systems. Two conventional cage
systems, one organic aviary system and one floor housing system were included. On average,
a higher (P < 0.001) initial eggshell contamination with total count of aerobic bacteria was
found for eggs from non-cage systems compared to conventional cage systems; respectively
5.46 compared to 5.08 log CFU/eggshell. However, initial contamination with total count of
Gram-negative bacteria on the eggshells was significantly lower (P < 0.001) in the non-cage
systems; 3.31 compared to 3.85 log CFU/eggshell. This study showed that the major
differences in eggshell contamination with total count of aerobic bacteria, found between
conventional and non-cage systems in the experimental studies (>1 log) were less
pronounced in the sampled commercial housing systems. The even lower initial
contamination with Gram-negative bacteria in the commercial non-cage systems was
remarkable.
e) On-farm studies
Six flocks of laying hens in furnished cages and seven flocks in non-cage systems (three
aviaries and four floor systems) were compared in the international study of De Reu et al.
(2009b). On average, eggshells from furnished cages were slightly, but significantly (P <
0.001), less contaminated with total count of aerobic bacteria compared to non-cage eggshells
(4.75 versus 4.98 log CFU/eggshell). In the non-cage systems, no difference in average
contamination between aviary and floor systems was found. Both within the groups of
furnished cage- and non-cage systems, major differences between farms were obtained.
Differences in farm management can possibly explain this. For Enterobacteriaceae no
significant difference in average eggshell contamination was found between furnished and
non-cage systems. Hunau-Salaun et al. (2009) also found a comparable higher eggshell
contamination for eggs from non-cage systems (4,82 CFU/eggshell) compared to
conventional cages (4,40 CFU/eggshell). On the other hand, no difference between furnished
cages and non-cage systems was found.
f) Bacterial air contamination and its relationship with eggshell contamination
In some studies the total count of aerobic bacteria in the air of poultry houses was proven to
be positively correlated with the initial bacterial eggshell contamination at the henhouse
(Protais et al., 2003a; De Reu et al., 2005b). Averages of 4 log CFU/m3 air for the
conventional and furnished cages were found compared with a 100 times higher average (> 6
log CFU/m3) for aviary housing systems.
g) Influence of housing system on quality of eggs and egg products
At this moment, it remains unknown whether the differences in bacterial counts on the shell
of eggs produced in different housing systems have an impact on the quality of eggs and egg
products. Harry (1963), De Reu et al. (2006b; 2006d) and many other researchers found a
correlation between bacterial eggshell contamination and egg infection or egg content
contamination. The higher prevalence of coliforms on the shells of eggs laid in furnished
cages was not correlated with signs of coliform contamination in egg yolk or albumen
(Cepero et al., 2000; Cepero et al., 2001). In a preliminary study of De Reu et al. (2007a;
2008), egg content contamination of nest eggs was 1.9% (5/269 eggs) for furnished cages
compared to 2.3% (10/432 eggs) for non-cage systems.
76
Aust. Poult. Sci. Symp. 2010...21
III. EGGSHELL DIRT AND CRACKS IN DIFFERENT HOUSING SYSTEMS
a) Eggshell dirt
Beside bacterial eggshell contamination the occurrence of eggshell dirt may also be
considered as a hygiene parameter. Additionally some types of eggshell dirt may give
nutrients to bacteria present on the shell. In table 1 the occurrence of dirty eggs in different
housing systems, found by different research groups, is summarized.
Table 1.
Occurrence of dirty eggs (in % of eggs) in different housing systems
Study
Type of study
Tauson et al. 1999
Mallet et al. 2006
De Reu et al. 2009b*
De Reu et al. 2009a
Pilot
Pilot
Comm.
Comm. (shop)
CC (%)
FC (%)
6.5
n.a.
4.9 (4.9-4.9) 5.0 (3.0-7.1)
n.a.
22
17.2
n.a.
Non-cage
(%)
5.7
n.a.
24
4.4
P
< 0.001
> 0.05
n.d.
CC = conventional cage, FC = furnished cage, Comm = commercial housing *= only nest eggs; n.a. =
not analyzed; n.d. not determined
Both in the commercial studies of De Reu et al. (2009a, 2009b) as in the pilot study of
Tauson et al. (1999) it was found that the frequency of dirty eggs in nests of non-cage
systems was not higher than in cage systems. On the other hand the study of Mallet et al.
(2006) showed that furnished cages can contain more dirty eggs compared to conventional
cages. Two types of conventional cages and two types of furnished cages were compared.
The study showed that a less optimal cage design of the furnished cages increased the number
of outside nest eggs which contained more eggshell dirt compared to the nest eggs. As a
result a significant difference was found in occurrence of dirty eggs between both types of
furnished cage designs; 3.0 and 7.1% respectively (Table 1). Comparing both conventional
cage designs with both furnished cage designs, on average both types of housing systems
contained a comparable amount of dirty eggs (4.9% compared to 5.0%). To conclude the
available research results indicate that nest eggs of non-cage systems are in normal
circumstances not more susceptible for dirtiness compared to cage eggs.
An important aspect is also that 85 up to 98% of the floor eggs of non-cage systems
have dirty eggshells (Abrahamsson and Tauson, 1998, De Reu et al. 2006a). This stresses the
fact that floor eggs are unfit as table eggs.
b) Eggshell cracks
As eggshell cracks give the opportunity for bacteria to penetrate the eggshell and hence
contaminate the egg content, the occurrence of cracks in eggs is also an important factor in
comparing housing systems. In table 2 the results of different research groups studying the
influence of housing systems on eggshell cracks are summarized.
The study of Guesdon et al. (2006) as well as the study of De Reu et al. (2009b)
showed that furnished cages are most susceptible for cracks. Guesdon et al. (2006) found in
pilot studies 15.4 to 19.6% of broken (visual observation) and hair-cracked (candling) eggs in
furnished cages compared to only 8.1 to 12.2% in standard cages. This was mainly due to
hair-cracked eggs at the narrow nests of the furnished cages without egg saver and a
relatively low frequency of manual egg collection. The study of De Reu et al. (2009b)
documented variations of cracked eggs from 0 up to 24% for the individual farms. A high
percentage (24%) of cracks was found in a furnished cage flock and was probably caused by
a bad adjustment of the egg saver and the accumulation of eggs next to the nest box on a
77
Aust. Poult. Sci. Symp. 2010...21
short part of the egg belt. In the study of Tauson et al. (1999) cracks varied from 2.2% to
7.7%, with no significant difference between the housing systems (conventional and floor
housing systems) and an almost comparable mean % of cracked cage eggs (5.0% compared
to 4.6%). In a market study on the quality characteristics of eggs from different housing
systems, Hidalgo et al. (2008) found no significant difference (P > 0.05) in appearance of
cracked eggs between cage (14%), free range (10%), barn (11%) and organic (5%) eggs. The
comparison at the shop level of De Reu et al. (2009a) also showed that non-cage eggs do not
contain more cracks compared to cage eggs (5.6 versus 7.8%). Of course eggs in those latter
studies were already candled and sorted at the packaging station. In summary the different
research results indicate that nest eggs of non-cage systems are in normal circumstances not
more susceptible for cracks compared to cage eggs. In addition, results indicate that a good
egg collection of the eggs from furnished cages is important to reduce cracks.
Table 2.
Occurrence of cracked eggs (in % of eggs) in different housing systems
Study
Tauson et al. 1999
Guesdon et al. 2006
De Reu et al. 2009b
Hidalgo et al. 2008
De Reu et al. 2009a
Type of study
Pilot
Pilot
Comm.
Comm. (shop)
Comm. (shop)
CC (%)
5.0
8.1 - 12.2
n.a.
14
7.8
FC (%)
n.a.
15.4 - 19.6
7.8%*
n.a.
n.a.
Non-cage (%)
4.6
n.a.
4.1
8.7
5.6
P
< 0.001
< 0.05
> 0.05
n.d.
CC = conventional cage, FC = furnished cage, Comm = commercial housing *= only nest eggs; n.a. =
not analyzed; n.d. not determined
IV. EFFECT OF HOUSING SYSTEM ON THE SALMONELLA CONTAMINATION
a) Salmonella contamination of eggs
Much of the research on eggshell and egg content contamination focuses on Salmonella,
since infection with Salmonella Enteritidis, resulting from the consumption of contaminated
eggs or egg products, is still a major health problem. Several EU member states have reported
data from investigations of table eggs, and the overall EU prevalence in 2006 was 0.8%
(EFSA, 2007). More than 90% of all egg-isolates were strains of the serotype Enteritidis.
Little research is done on the influence of housing system on eggshell and egg content
contamination with Salmonella. In a study of Humphrey et al. (1991), over 5700 eggs from
15 naturally infected flocks were examined, of which 32 or 0.6% were contaminated. The
prevalence of egg content contamination of eggs from battery or free-range were comparable;
0.73 and 0.64% respectively. A study of the UK Food Standards Agency in 2003 also did not
found significant differences in Salmonella spp. contamination on the shell due to the
production system (Anon., 2004). On a total of 4753 retail samples of boxes with six eggs,
the eggshell of nine samples was contaminated. None of the 4753 pooled egg contents of
retail samples were Salmonella positive. In a smaller study of De Reu et al. (2009a) 47 fresh
egg samples from the Belgian market were sampled. Sixteen samples concerned cage-eggs,
five floor housing eggs, 12 free range eggs, seven organic eggs, five samples from farm retail
and two from private backyards. In none of the samples Salmonella was found.
b) Environmental Salmonella contamination
More research was focused on the influence of the housing system on environmental
Salmonella contamination. The analysis of the existing data of an EU-wide baseline study on
Salmonella in laying hens flocks performed in 2004-2005 (EFSA 2006) showed a significant
78
Aust. Poult. Sci. Symp. 2010...21
difference in Salmonella prevalence according to housing type (Anon. 2008). In cage systems
the highest Salmonella prevalence was found followed by a intermediate prevalence in barn
systems and the lowest prevalence in the free range systems. More than 51% of all
Salmonella isolates were serotyped as Salmonella Enteritidis. Recent data collected in the
large-scale cross-sectional and longitudinal field study of the EU Safehouse project confirm
those findings (Anon. 2008). An overview by Dewulf et al. (2009), on published
observational studies evaluating the effect of housing system on the prevalence of Salmonella
Enteritidis infections, also clearly indicates that a cage system has an increased risk for being
Salmonella positive in comparison to non-cage housing systems. However, there is not
necessarily a causal relationship between the housing type and the Salmonella infection. It is
more likely that the housing system is a proxy of many other production characteristics such
as magnitude of the flock or herd, age of the building, probability of previous Salmonella
infection on the farm. The authors summarized a number of important production
characteristics that may be both related to the housing system and the probability of a
Salmonella infection: herd and flock size, stocking density, stress, age of the building and
carry-over infections, pests, vaccination.
V. CONCLUSIONS
It is clear that eggshell contamination with aerobic bacteria is on average significantly higher
for nest eggs from non-cage systems compared to nest eggs from furnished cages or eggs
from conventional cages. The major differences found in experimental studies between cage
and non-cage systems are less pronounced under commercial circumstances. The scarce
information available on the influence of the housing systems on the egg content
contamination indicates no major differences in egg content contamination between cage
eggs and non-cage eggs (ignoring outside nest and floor eggs).
Studies also show that it is highly unlikely that a move from conventional cages to
alternative cage systems and non-cage housing systems will result in an increase in
Salmonella infection and shedding, rather the opposite is expected.
VI. ACKNOWLEDGEMENT
The cited research of our group would not have been possible without the help of especially
Ann Van de Walle. Sofie De Vlam, Elly Engels and Vera Van de Mergel are also gratefully
acknowledged.
REFERENCES
Abrahamsson P, Tauson R (1998) Performance and egg quality of laying hens in an aviary
system. Journal of Applied Poultry Research 7, 225-232.
Anon. (2004) Report of the survey of Salmonella contamination of UK produced shell eggs
on retail sale. London, Food Standard Agency, 124 p.
Anon. (2008) Newsletter 1. Safehouse and Rescape. [F Van Immerseel, Y Nys Eds.]
Cepero R, Yangüela J, Lidon MD, Hernandis A (2000) Conference proceeding - XXXVII
Symposium Sec. Esp. WPSA, I Congreso Internacional de Sanidad y Producción
Animal, pp 61-80. Barcelona, Spain.
Cepero R, Maria G, Hernandis A (2001) Conference proceedings - XXXVIII Symposium Sec.
Esp. WPSA, I Congreso Internacional de Sanidad y Producción Animal, Córdoba,
Spain.
79
Aust. Poult. Sci. Symp. 2010...21
De Reu K, Grijspeerdt K, Heyndrickx M, Uyttendaele M, Herman L (2005a) The use of total
aerobic and Gram-negative flora for quality assurance in the production chain of
consumption eggs. Food Control 16, 147-155.
De Reu K, Grijspeerdt K, Heyndrickx M, Zoons J, De Baere K, Uyttendaele M, Debevere J,
Herman L (2005b) Bacterial eggshell contamination in conventional cages, furnished
cages and aviary housing systems for laying hens. British Poultry Science 46, 149155.
De Reu K (2006a) Bacteriological contamination and infection of shell eggs in the production
chain. PhD Thesis, Faculty of Bioscience Engineering, University of Ghent, Belgium.
De Reu K, Grijspeerdt K, Heyndrickx M, Messens W, Uyttendaele M, Debevere J, Herman L
(2006b) Influence of eggshell condensation on eggshell penetration and whole egg
contamination with Salmonella enterica serovar Enteritidis. Journal of Food
Protection 69, 1539-1545.
De Reu K, Grijspeerdt K, Heyndrickx M, Uyttendaele M, Debevere J, Herman L (2006c)
Bacterial eggshell contamination in the egg collection chains of different housing
systems for laying hens. British Poultry Science 47, 163-172.
De Reu K, Grijspeerdt K, Messens W, Heyndrickx M, Uyttendaele M, Debevere J, Herman L
(2006d) Eggshell factors influencing eggshell penetration and whole egg
contamination by different bacteria, including Salmonella Enteritidis. International
Journal of Food Microbiology 112, 253-260.
De Reu K, Heyndrickx M, Grijspeerdt K, Rodenburg B, Tuyttens F, Uyttendaele M, Herman
L (2008) Estimation of the vertical and horizontal bacterial infection of hen’s table
eggs. World’s Poultry Science Journal, 64, 142.
De Reu K, Renders K, Maertens G, Messens W, Reybroeck W, Ooghe S, Herman L,
Daeseleire E (2009a) A market study on the quality of eggs from different housing
systems. Proceeding on XIXth European Symposium on the Quality of Poultry Meat &
XIIIth Symposium on the Quality of Eggs and Egg Products. Turku, Finland.
De Reu K, Rodenburg B, Grijspeerdt K, Heyndrickx M, Tuyttens F, Sonck B, Zoons J,
Herman L (2009b) Bacteriological Contamination, Dirt and Cracks of Eggshells in
Furnished Cages and Non-Cage Systems for Laying Hens, An International On-Farm
Comparison. Poultry Science 88, 2442-2448.
Dewulf J, Van Hoorebeke S, Van Immerseel F (2009) Epidemiology of Salmonella infection
in laying hens with special emphasis on the influence of the housing system.
Proceeding on XIX th European Symposium on the Quality of Poultry Meat & XIIIth
Symposium on the Quality of Eggs and Egg Products. Turku, Finland.
EFSA, European Food Safety Authority (2006) Preliminary report on the analysis of the
baseline study on the prevalence of Salmonella in laying hen flocks of Gallus gallus.
The EFSA Journal 81, 1-71.
EFSA, European Food Safety Authority (2007) The Community summary report on trends
and sources of zoonoses, zoonotic agents, antimicrobial resistance and foodborne
outbreaks in the European Union in 2006.
Gentry RF, Quarles CL (1972) The measurement of bacterial contamination on eggshells.
Poultry Science 51, 930-933.
Guesdon, V., Ahmed, A. M. H., Mallet, S., Faure, J.M., and Nys, Y. (2006). Effects of beak
trimming and cage design on laying hen performance and egg quality. British Poultry
Science 47,1-12.
Harry EG (1963) The relationship between egg spoilage and the environment of the egg when
laid. British Poultry Science 4, 91-100.
80
Aust. Poult. Sci. Symp. 2010...21
Humphrey TJ, Whitehead A, Gawer AHL, Henley A, Rowe B (1991) Numbers of Salmonella
enteritidis in the contents of naturally contaminated hen's eggs. Epidemiology and
Infection 106, 489-496.
Huneau-Salaun A, Michel V, Huonnic D, Balaine L, Le Bouquin S (2009) Factors
influencing bacterial eggshell contamination in conventional cages, furnished cages
and free-range systems for laying hens under commercial conditions. British Poultry
Science, in press.
Hidalgo A, Rossi M, Clerici F, Ratti S (2008) A market study on the quality characteristics of
eggs from different housing systems. Food Chemistry 106,1301-1308.
Mallet S, Guesdon V, Ahmed AMH, Nys Y (2006) Comparison of eggshell hygiene in two
housing systems: Standard and furnished cages. British Poultry Science 47, 30-35.
Messens W, Grijspeerdt K, Herman L (2005) Eggshell characteristics and penetration by
Salmonella enterica serovar Enteritidis through the production period of a layer
flock. British Poultry Science 46, 694-700.
Protais J, Queguiner S, Boscher E, Piquet J-C, Nagard B, Salvat G (2003a) Effect of housing
system on the bacterial flora in the air and on egg shells, Conference proceedings - Xth
European Symposium on the Quality of Eggs and Egg Products, 142-149. SaintBrieuc, Ploufragan, France,
Quarles CL, Gentry RF, Bressler GO (1970) Bacterial contamination in poultry houses and
its relationship to egg hatchability. Poultry Science 49, 60-66.
Rodenburg B, Tuyttens F, De Reu K, Herman L, Zoons J, Sonck B (2008) Welfare
assessment of laying hens in furnished cages and non-cage systems, an on-farm
comparison. Animal Welfare 17,363-373.
Tauson R, Wahlström A, Abrahamsson P (1999) Effect of two floor housing systems and
cages on health, production, and fear response in layers. Journal of Applied Poultry
Research 8,152-159.
Wall H, Tauson R, Sorgjerd S (2008) Bacterial contamination of eggshells in furnished and
conventional cages. Journal of Applied Poultry Research 17,11-16.
81
Aust. Poult. Sci. Symp. 2010...21
LITTER CONSUMPTION BY POULTRY AS AFFECTED BY DIET STRUCTURE
B. SVIHUS 1 and H. HETLAND1
Summary
Based on previous results indicating that poultry voluntarily consume litter material, two
experiments were carried out to investigate the effect of access to litter for broiler chickens
with diets with varying content of structural components. The results indicate that birds
consume litter material, and that the amount consumed is inversely related to amount of
structural components in the diet.
I. INTRODUCTION
It has been shown in numerous experiments that structural components such as whole wheat
and oat hulls will stimulate gizzard development, and that the resulting more muscular and
voluminous gizzard may improve nutrient digestibility and performance (Svihus et al., 2004;
Ravindran et al., 2006; Gabriel et al., 2008). Although the mechanism for the beneficial effect
has not been fully revealed, data indicate that the beneficial effect is associated with a longer
retention time and a lower pH of the gizzard, and a finer grinding of the diet ingredients
(Gabriel et al., 2003; Hetland et al., 2002).
It has also been shown that birds will voluntarily consume fibrous components presented
for the birds in a different trough than the feed, and that such a consumption of fibrous
components is dependent on the coarseness of the diet (Hetland et al., 2005). This indicates
that birds have a desire for structural components, and that birds will search for structural
components which can be consumed in a situation where the diet lacks structure. Litter
material used as bedding in floor systems may be such an alternative source for structural
components. Thus, we have carried out two experiments to study litter consumption of
broilers.
II. MATERIALS AND METHODS
In study one, day-old male broiler chickens (Ross) were reared in cages with mesh-wire floor
on commercial broiler starter feed till 11 days of age. From 6 days of age, 50 g/kg of whole
wheat was mixed into the starter diet for half of the flock to allow the birds to get used to the
whole wheat diet. At 11 days of age 180 chickens were randomly and equally distributed
among 12 pens (75 cm x 150 cm). A constant daily temperature (28 C° from 11-14 days and
26 C° thereafter) and 23 hours light was provided. Half the pens were supplied with 1.5 kg
litter, while the other half was supplied with a green rubber mesh mat. The litter was sieved
through a 5 mm sieve, and only particles larger than 5 mm were used. All birds received the
same wheat-based diet (779 g/kg), but half the birds received a pelleted diet with 500 g/kg
whole wheat added before pelleting while the other half received a pelleted diet based on
wheat ground through a 3 mm hammer mill before pelleting. The diets and water were given
ad libitum. Chickens and feed were weighed at 14, 21 and 28 days of age. At termination of
the experiment, all the litter material was soaked in water for 12 hours, followed by rinsing
(10 litre per minute) through a 5 mm sieve and thereafter drying of the material remaining on
the sieve for 3 days at 105 C°.
1
Norwegian University of Life Sciences, P.O. Box 5003, N-1432 Aas, Norway
82
Aust. Poult. Sci. Symp. 2010...21
In study two, a total of 360 broiler chickens (Ross 308) were placed in 24 pens with
15 birds per pen at 4 days of age. One half of the pens had rubber mat on the floor, and the
other half had litter on the floor. One half of the pens on each floor type were fed on a
pelleted wheat-based (709 g/kg) diet and the other half on the same diet but diluted with 5%
coarse oat hulls before pelleting. This represents a 2 x 2 factorial design with the factors
being dietary fibe (low and high) and litter (absent and present). Birds and feed were weighed
at the start, day 19, and day 32.
Table 1.
Results from Experiment 1
Wheat
treatment
Weight gain 11-29
Ground
days, g
Whole
Feed consumption 11- Ground
29 days, g
Whole
Feed/gain
Ground
Whole
Ileal starch
Ground
digestibility, %
Whole
Empty gizzard, % of
Ground
live weight
Whole
Gizzard content, % of
Ground
live weight
Whole
Fibre concentration in Ground
gizzard, %1
Whole
No
litter
1175
1405
1794
1922
1.64
1.70
89
88
1.4
1.6
0.3
0.7
17.2
34.4
Litter
1267
1226
1723
1713
1.48
1.43
94
90
1.5
1.7
0.6
0.8
37.0
38.7
P-value for
litter effect
0.6836
P-value for
diet effect
0.3441
P-value for
interaction
0.4464
0.0391
0.1558
0.0960
0.1054
0.7245
0.9812
0.0333
0.1156
0.8946
0.0204
0.0004
0.7039
0.0032
0.0001
0.2858
0.0001
0.0030
0.1826
1
Based on the NDF analysis (Van Soest method).
III. RESULTS AND DISCUSSION
Weight gain was not significantly affected by either structural components in the diet or
whether the birds were raised on wood shavings litter or not. The lack of bedding effect
indicates that both environmental conditions were suitable for the birds. However, access to
litter had a significant moderating effect on feed intake, which in turn resulted in a
significantly improved feed/gain for birds with access to litter in experiment 2. As ileal starch
digestibility was significantly improved with access to litter in experiment 1, this indicates
that the cause for improved feed utilization was at least partly an improved nutrient
digestibility.
Since gizzard size, gizzard content weight and fibre concentration increased for birds
raised on litter, it is logical to conclude that birds consumed litter. In experiment one,
weighing of the litter for each cage before and after the experiment indicated that birds were
eating approximately one gram per bird per day. Since it has been shown in several
experiments carried out previously that structural components in the diet stimulate gizzard
development and starch digestibility, it can therefore be postulated that litter consumption is
the cause for increased gizzard size, and thus the primary cause for increased starch
digestibility and reduced feed/gain. Although addition of hulls to the diet resulted in a
significantly reduced size of particles in the duodenum, in accordance with results observed
before (Hetland et al., 2002), access to litter did not have the same effect. This could be due
to a lower magnitude of litter structural effect caused by moderate consumption, as observed
in experiment 1. The significant trend towards a larger increase in gizzard size for diets
83
Aust. Poult. Sci. Symp. 2010...21
without hulls in experiment 2 indicates that consumption of litter material interacts with
access to structural components through the diet. This lends support to the hypothesis put
forward earlier, that poultry eat litter material to compensate for lack of structural
components in the diet (Hetland et al., 2004; Hetland et al., 2005). Thus, these experiments
indicate that litter material, when used for birds given a modern broiler diet with a low
content of structural material, should be regarded not only as a bedding material, but also as a
dietary ingredient. This insight has bearing for the hygienic conditions of the digestive tract,
and it has bearings for the choice of litter materials.
Table 2.
Results from Experiment 2
Diet
Weight gain 6-32
days, g
Feed consumption 632 days, g
Feed/gain
Starch content in
ileum, %
Empty gizzard
weight, g on day 19
Empty gizzard
weight, g on day 32
Mean duodenal
particle µm on day 19
Mean duodenal
particle µm on day 32
No hulls
Hulls
No hulls
Hulls
No hulls
Hulls
No hulls
Hulls
No hulls
Hulls
No hulls
Hulls
No hulls
Hulls
No hulls
Hulls
No
litter
1957
1949
3181
3021
1.63
1.55
12.2
1.1
15.5
25.6
27.6
44.9
160
89
237
136
Litter
1919
1960
2945
2931
1.53
1.50
9.6
1.3
21.9
26.3
31.0
43.4
120
110
264
172
P-value for
litter effect
0.6802
P-value for
diet effect
0.6205
P-value for
interaction
0.4486
0.0136
0.1623
0.2376
0.0006
0.0044
0.4469
0.1042
<0.0001
0.0622
0.0037
<0.0001
0.0169
0.5303
<0.0001
0.0939
0.6821
0.0748
0.1809
0.0908
<0.0001
0.8169
REFERENCES
Gabriel I, Mallet S, Leconte M, Travel A, Lalles JP (2008) Animal Feed Science and
Technology 142, 144-162.
Gabriel I, Mallet S, Leconte M (2003) British Poultry Science 44, 283-290.
Hetland H, Svihus B, Olaisen V (2002) British Poultry Science 43, 416-423.
Hetland H, Choct M, Svihus B (2004) World’s Poultry Science Journal 60, 415-422.
Hetland H, Svihus B, Choct M (2005) Journal of Applied Poultry Research 14, 38-46.
Ravindran V, Wu YB, Thomas DG, Morel PCH (2006) Australian Journal of Agricultural
Research 57, 21-26.
Svihus B, Juvik E, Hetland H, Krogdahl Å. (2004) British Poultry Science 45, 1-6.
84
Aust. Poult. Sci. Symp. 2010...21
THE EFFECTS OF INTERRUPTING A DUSTBATHING BOUT ON THE CHOICE
BEHAVIOUR OF LAYING HENS IN A Y-MAZE TEST
S.M. LAINE 1, G.M. CRONIN 2, J.C. PETHERICK 3 and P.H. HEMSWORTH1
It has been suggested that interrupting the interaction of an animal with the reward provided
in a choice test can be aversive and devalue the reward (e.g. Mason et al., 1998). Previous
work of ours with hens given access to a dustbathing substrate for different lengths of time in
a Y-maze preference test appeared to support this suggestion; hens allowed 20 minutes of
interaction with peat moss appeared less motivated to choose peat moss on subsequent trials
than hens given access for 45 minutes or 2 minutes (Laine et al., 2009). The average duration
of a dustbathing bout of a laying hen is 27 minutes (Vestergaard, 1982), however, the hens
receiving 20 minutes of dust reward in Laine et al. (2009) had an average bout duration of
only 15 minutes. To determine if interrupting dustbathing mid-way through a bout is
aversive, the preference of hens in a Y-maze was investigated when they were offered
uninterrupted dustbathing vs. dustbathing interrupted after 15 minutes (unint vs. 15) and
dustbathing interrupted after 15 minutes vs dustbathing interrupted after 2 minutes (15 vs. 2).
For each hen the arm in which the options were presented were held constant and the arm
visually cued (with a black or white board) at the choice point in the Y-maze. Sixteen hens
were preference tested eight times, one trial per day on alternate days and were given 15
minutes in which to commence dustbathing after making a choice. The choice made
determined whether dustbathing was interrupted and after what time. Each hen underwent
both treatments over two periods (eight hens per treatment per period). Choice behaviour was
compared to chance level (i.e. 50:50) by a Chi-square test and this revealed that the hens
showed no preference on all trials (treatments unit vs. 15 and 15 vs. 2, P = 0.69 and P = 0.42
respectively) or on trials when dustbathing occurred (treatments unit vs. 15 and 15 vs. 2, P =
1.0 and P = 0.23 respectively). As a lack of preference was unexpected, we conducted a
second experiment to determine whether the lack of preference was just that or an inability to
learn the association between the visual cue, maze arm and duration of dustbathing. Thirteen
hens were preference tested in a Y-maze for their choice between a dustbath filled with peat
moss or an empty dustbath, which was visually cued with a black or white board as in
experiment 1. Hens could not see the dustbath contents until they made their choice. As
previously, for each hen the arms in which the rewards were presented were held constant.
Hens were given eight trials (one per day) on alternate days. The hens chose the dust-filled
dustbath on significantly more trials than the empty dustbath (P < 0.001) indicating that hens
can learn a simple task of associating a visual cue and arm maze with access to a substrate.
Even though the hens were successful at learning the task in experiment 2, it is possible that
hens were unable to learn the complex task in the first experiment. Alternatively, contrary to
our hypothesis, interruption of dustbathing mid-way through a bout was not aversive.
Laine SM, Cronin GM, Petherick JC, Hemsworth PH (2009) Proc. Aust. Poult. Sci. Symp. 21,
153-156.
Mason G, McFarland D, Garner J (1998) Anim. Behav. 55, 1071-1075.
Vestergaard K (1982) Appl. Anim. Ethol. 8, 487-495.
1
Animal Welfare Science Centre, Melbourne School of Land and Environment, University of
Melbourne. Parkville Vic 3010.
2
Faculty of Veterinary Science, University of Sydney, Camden NSW 2570
3
Queensland Primary Industries and Fisheries, Department of Employment, Economic Development
and Innovation, Rockhampton Qld 4702
85
Aust. Poult. Sci. Symp. 2010...21
EGG QUALITY IN THE AUSTRALIAN EGG INDUSTRY: AN UPDATE
J.R. ROBERTS 1 and K.K CHOUSALKAR 2
Summary
Eggs were sampled from commercial egg farms in NSW, QLD, VIC, SA and WA. A total of
6300 eggs were analysed on-site for egg weight, albumen height and Haugh Units. In
addition, a total of 18,039 eggs were analysed in the laboratory for the full range of egg shell
quality and egg internal quality measurements. Data were compared for state of Australia,
strain of bird and production system. In general, egg shell quality and egg internal quality
were relatively independent of state, strain of bird and egg production system although there
was a range of values for all parameters measured.
I. INTRODUCTION
The egg is the final product of the Australian Egg Industry and its internal and external
quality is of paramount importance to the industry and the consumer. Quality Assurance
programs are an essential feature of all egg producing establishments. The current project
conducted egg quality testing, further to the earlier AECL study (Roberts & Ball, 2004) to
enable the continuing development of a data base of egg internal quality and egg shell quality
within the Australian Egg Industry. In recent years, there have been some acute problems
with poor albumen quality. The current project specifically targeted albumen quality of
freshly-laid eggs taken straight from the cage front.
Australian per capita consumption of eggs (196 as at September, 2009) is about one
third of egg consumption in the USA and Canada. Therefore, the potential exists for a
substantial increase in egg consumption in Australia, particularly now that the AECL has
been successful in achieving the Heart Foundation’s Tick of Approval for eggs.
Losses of eggs owing to poor shell quality have been conservatively estimated at
10%. For the Australian industry (13 million hens x 27 dozen eggs at a farm-gate price of
$1.60 per dozen) each 1% loss in saleable eggs is approximately $5 million annually.
However, losses in saleable eggs would also include the input costs (pullet, feed) associated
with producing that egg (analysis by TA Scott, 2006).
II. MATERIALS AND METHODS
The project involved egg quality testing from a range of poultry establishments for the
purposes of updating the existing AECL egg quality database. Research and industry contacts
in most states of Australia sampled eggs, 30 per flock, directly from the cage front for
measurement of albumen quality. A sample of 90 eggs from the same source was sent by
courier to the University of New England where they were subjected to the full range of egg
quality tests: shell colour, shell breaking strength and deformation, shell weight, shell
thickness, albumen height, Haugh Units, yolk colour score. Routine sampling was conducted
across a range of flock ages, including early, mid and late lay. Sampling was largely
opportunistic, depending on which producers agreed to participate in the study and were
1
2
School of Environmental and Rural Science, Univ of New England, Armidale, NSW 2351
School of Animal and Veterinary Sciences, Charles Sturt Univ, Wagga Wagga, NSW 2650
86
Aust. Poult. Sci. Symp. 2010...21
geographically accessible to the project sampling team. Egg internal quality was measured
on-farm as egg weight and albumen height and Haugh Units calculated. In the laboratory, egg
quality was measured as shell reflectivity, egg weight, deformation, breaking strength, shell
weight (form which percentage shell was calculated), shell thickness, egg length and breadth
(from which shape index was calculated as breadthx100/length). Data were analysed by
ANOVA and Fisher’s protected LSD was used to distinguish differences between means.
Significance was assumed at P < 0.05.
III. RESULTS
A total of 6300 eggs were analysed on-site for egg weight, albumen height and Haugh Units.
In addition, a total of 18,039 eggs were analysed in the laboratory for the full range of egg
shell quality and egg internal quality measurements. Eggs were collected from 210 flocks in
total, 67 from NSW, 54 from VIC, 55 from QLD, 20 from SA, 14 from WA. Of these, 156
flocks were cage production, 46 free range and 8 barn. Strain distribution was HyLine 114
flocks, Isa 52, HiSex 44 flocks. Most flocks were single age and multi-age flocks were
reported for only 29% of cage flocks. Reticulated town water was the water source for most
of the flocks (93) followed by bore water (78) and dam (29). The method used to treat water
coming from non-town sources was predominantly a combination of filtration and
chlorination with some farms using reverse osmosis or iodine treatment. For cage production,
most single age sheds had controlled ventilation and most multi-age cage flocks came from
older style sheds with natural ventilation. Free range and barn flocks had shedding that was
mainly naturally ventilated although some had some degree of ventilation control. Years of
staff experience varied from less than one year to 60 years although the averages were similar
for cage (22 yrs) and barn (23 years) which were higher than for free range (16 yrs). Most
farms collected eggs at least daily although 17% of eggs from both cage and barn systems
were collected only 6 days per week. The average temperature of egg storage rooms was
14°C for all production systems but varied between 9 and 17°C. Three quarters of cage flocks
and approximately half of all free range and barn flocks were vaccinated for infectious
bronchitis (IB) virus only during rearing. For flocks that were revaccinated regularly during
lay, the frequency of revaccination generally ranged between 6 and 10 weeks. Relatively few
cage and free range flocks were reported as having shown signs of IB infection but this
proportion was higher for barn flocks. The majority of flocks were audited regularly, most by
AECL Egg Corp Assured (ECA). Most farms used cardboard fillers. Age at first egg was
similar for all production systems (17.5 weeks) with peak production occurring at 27-30
weeks of age and 90% at 44, 51 and 55 weeks of age for free range, cage and barn,
respectively.
The effects of flock age were very similar to those which have been reported earlier
(Roberts & Ball, 2004). Egg weight increased and then stabilised at between 60 and 70
grams. Shell reflectivity increased with hen age initially but then remained relatively stable.
Shell deformation generally decreased with hen age and shell breaking strength decreased
with hen age. Shell weight generally increased in a manner similar to egg weight but, for
some flocks, tended to decrease later in lay. Shell thickness was relatively stable across a
range of hen ages. Percentage shell was relatively constant until 60 weeks of age after which
it tended to decrease. Albumen height and Haugh Unit generally decreased with increasing
age of the flock and were more variable for the values measured in the laboratory, owing to
varying lengths of time elapsing between egg collection and measurement. Yolk colour
varied between 8 and 13 and was independent of hen age (although two barn flocks from
VIC were well below the average).
87
Aust. Poult. Sci. Symp. 2010...21
There was generally no difference among states for the egg quality variables
measured: egg weight; albumen height and Haugh Unit (both on-farm and laboratory); shell
reflectivity (although three free range flocks from SA had significantly higher shell
reflectivity); shell deformation; shell breaking strength; shell weight; shell thickness
(although some individual flocks in NSW were below average); shell thickness; percentage
shell; yolk colour.
The relationship between egg weight and hen age was similar for all strains.
Albumen height measured on freshly collected eggs declined with hen age for all strains but
was generally highest for the HyLine Brown and lowest for the HiSex birds, with Isa
intermediate. Haugh Unit, which takes into account the size of the egg, declined with hen
age but was more similar among strains of bird than was albumen height for freshly
collected eggs. There was a greater variability in both albumen height and Haugh Unit
measured in the laboratory owing to different ages of the eggs. There was considerable
overlap among strains for shell reflectivity. Three free range flocks, two HyLine and one
HiSex had significantly lighter shell colour. Shell breaking strength showed a considerable
degree of overlap, but some flocks were noticeably below the average. Shell weight was
generally highest for ISA and lowest for HyLine with HiSex intermediate. There was no
consistent difference among strains for shell thickness and percentage shell was generally
similar for all strains. Albumen height and Haugh Unit measured in the laboratory showed
considerable variation because of the different amounts of time that had elapsed between the
collection of the eggs and their analysis in the laboratory. For yolk colour, two barn flocks
from VIC were well below the average.
There were no differences among production systems for egg weight; albumen
height and Haugh Unit; shell breaking strength, shell thickness; percentage shell and yolk
colour score. However, there were two barn flocks which had very low yolk colour score.
Shell reflectivity was generally higher for free range flocks with several flocks well above
average. Shell deformation was similar for all production systems. Shell weight was
generally higher for ISA than for the other two strains. Yolk score was independent of hen
age and production system.
Albumen height was higher when measured on-farm than when measured at varying
time intervals later in the laboratory. Haugh Unit was also higher when measured on-farm
than when measured later in the laboratory. When the loss in Haugh Units with time
between on-farm analysis and analysis later in the laboratory are plotted on a graph, it can
be seen that, for most flocks, there is a loss of Haugh Unit but, for other flocks, Haugh Unit
was actually higher when measured in the laboratory. In general, the loss of Haugh Units
was between 1 and 10 up to 15 days between analyses and between 10 and 20 for longer
time delays between analyses.
IV. DISCUSSION
In general, egg shell quality and egg internal quality were relatively independent of state,
strain of bird and egg production system although there was a range of values for all
parameters measured. When flocks from different states were compared, several free range
flocks from SA had lighter coloured shells, some QLD flocks had higher shell deformation
and shell thickness and some NSW flocks were significantly below average for shell
thickness. When the three strains of bird were compared, there were some differences.
Albumen height of freshly measured eggs was generally highest for HyLine and lowest for
HiSex although much of this could be explained by differing egg weights and there was less
variation among strains for Haugh Unit. A small number of flocks, including three free
range flocks, had lighter coloured shells. The cause of this reduced pigmentation is not
88
Aust. Poult. Sci. Symp. 2010...21
known athough suggestions include anaemia (Juergen Lohr, personal communication). The
two barn flocks that had very low yolk colour appear not to have had pigment added to the
feed. There were relatively few differences among production systems. As expected,
albumen height and Haugh Unit measured later in the laboratory were generally lower than
those measured directly at the cage front. However, watery whites were encountered only
rarely and not consistently throughout a flock. This finding suggests that there is not a major
problem with water albumen in Australian layer flocks. Comparison of these results with
those obtained in 2003 reveal some general differences with egg weight being lower, egg
shell colour darker, shell deformation lower and shell thickness higher in the 2009 study.
When the results of this study are compared with those of the study published in
2003 (Roberts & Ball, 2004), egg weight in the 2009 study is generally lower than for the
2003 study. Egg shell colour measured in the 2009 study was generally darker (lower
reflectivity) than for the 2003 study, with the exception of the three free range flocks,
mentioned earlier. Shell deformation measured in the 2009 study tended to be lower than for
the 2003 study. Shell breaking strength and shell weight were very similar for the two
studies. Shell thickness tended to be higher for the 2009 study. Percentage shell was very
similar for the 2003 and 2009 studies. Albumen height measured in the laboratory in 2003
was generally higher than that measured in the laboratory in 2009. Albumen height
measured in the laboratory in 2003 and on-farm in 2009 was very similar for the two
studies. Haugh Unit measured in the laboratory in 2003 was similar to that measured in the
laboratory in 2009. Haugh Unit measured on-farm in 2009 tended to be higher than that
measured in the laboratory in 2003.
V. ACKNOWLEDGEMENTS
We thank all the producers who agreed to participate in this project for their generous
cooperation as well as our colleagues who assisted in the project, particularly Rowly Horn.
This study was supported by a grant from Australian Egg Corporation Limited which also
funded a Postdoctoral Fellowship for Kapil Chousalkar.
REFERENCES
Roberts JR, Ball W (2004) Egg Quality Guidelines for the Australian Egg Industry. AECL
Publication No. 03/19 (AECL Project UNE 71-A).
89
Aust. Poult. Sci. Symp. 2010...21
OPTIMAL SULPHUR AMINO ACIDS TO LYSINE RATIO IN GROWER PHASE
IN ROSS 308 BROILERS
T.G. MADSEN 1, E. HANGOOR 2, P.J.A. WIJTTEN2, J.K.W.M. SPARLA2 and A. LEMME 3
Summary
An experiment with growing male Ross 308 broilers (14-35 days) was performed to test the
optimal ratio between true fecal digestible sulphur amino acids (DSAA) and true fecal
digestible lysine (DSAA:DL). The ratio was increased from 63% to 82% with increments of
approximately 5%. Live weight gain and feed conversion ratio (FCR) was measured for the 3
week period. At day 35 the broilers were slaughtered and carcass and breast meat yield were
measured Results indicate that Ross 308 broilers respond positively to DSAA:DL ratio’s
higher than currently recommended and that optimum varies with performance parameter
with the highest optimum for FCR, carcass yield, and breast meat yield. Based on these data
it seems like the economic optimum will require a higher DSAA:DL ratio when the
production goal is cut up chicken compared to whole birds.
I. INTRODUCTION
The Ideal Protein (IP) concept is a tool enabling quick and easy adjustments to changing
production conditions. Thus, knowledge only of optimum dietary lysine levels for certain
conditions are needed while the ratios between all essential amino acids and lysine are
maintained the same. The optimal ratio between sulphur amino acids and lysine has been
studied in a number of trials (Mack et al., 1999; Vieira et al., 2004). However, the optimum
ratio is still being discussed Taherkhani et al. (2008). Recently Rostagno et al. (2007)
reported optimal ratios well above the NRC (1994) recommendations of 72%. It has also
been reported that requirements for sulphur amino acids depends on the parameters tested,
e.g. growth, feed conversion ratio (FCR), and carcass composition (Rostagno et al., 2007;
Lumkins et al., 2007). Therefore, an experiment with growing male Ross 308 broilers was
performed to test the effect of increased ratio between digestible sulphur amino acids
(DSAA) and digestible lysine (DSAA:DL) on growth, FCR, carcass yield and breast meat
yield.
II. MATERIAL AND METHODS
The experiment was conducted with 720 male Ross 308 broilers from 14 to 35 days of age.
The birds were all housed in 36 pens with raised floors at the Broiler Research Facility of
Provimi’s Research Centre in The Netherlands. The DSAA:DL ratio was increased from 63%
to 82% with increments of approximately 5% by adding DL-Methionine to the sulphur amino
acids deficient diet. An additional treatment was included in which the diet with highest
DSAA:DL ratio was supplemented with extra L-Lysine HCl in order to provide evidence that
digestible lysine was second limiting. Each of the 6 treatments was repeated six times and the
birds had ad libitum access to feed and fresh drinking water. Up to the start of the trial at day
14 the broiler chicks were fed a commercial starter diet. The diet and nutrient composition of
1
Health and Nutrition, Evonik Degussa (SEA) Pte. Ltd, Singapore
Provimi Research and Innovation Centre, Brussels, Belgium.
3
Health and Nutrition, Evonik Degussa GmbH, Hanau, Germany
2
90
Aust. Poult. Sci. Symp. 2010...21
the 63 % DSAA:DL ratio diets and the positive control diet (high lysine) is given in Table 1.
All diets were kept isocaloric and CP was also kept stable.
Live weight gain and FCR was measured for the 3 week period. At day 35 the broilers
were slaughtered and carcass and breast meat yield were measured. Data were analysed by
analysis of variance and the Tukey-Kramer test was applied to account for multiple
comparisons between treatments. The optimal ratio between DSAA and digestible lysine
(DL) was determined by exponential regression (y = a + b * (1-e(-c * x)) where the optimum
were set at 95% of asymptotic response (see Figure 1).
Table 1.
Diet and nutrient composition for the diet with low digestible sulphur amino
acids (MC) and digestible lysine ratio (low MC:L) and the positive control
(PC) diet with adequate lysine content.
Diet composition,
Treatments
Nutrient
Treatments
g/kg
Low MC:L
PC
composition, g/kg
Low MC:L*
PC
Maize
485.3
485.3
Crude protein
202
204
Wheat
50.0
50.0
Crude fat
104
103
Soybean meal
325.9
325.9
Crude fibre
24
24
Maize gluten
18.2
18.2
Crude ash
60
58
Soybeans
6.1
6.1
AMEn (kcal/kg)
3218
3220
Animal fat
35.9
35.7
TFD Lys
9.44
10.46
Soybean oil
35.9
34.8
TFD Met
3.02
4.86
Premix
10.0
10.0
TFD Met + Cys
5.94
7.78
Limestone
12.7
12.7
TFD Thr
7.32
7.32
MonoCaPO4
10.9
10.9
TFD Trp
2.14
2.14
Salt
2.3
1.8
TFD Ile
8.29
8.29
Sodium bicarbonate
1.5
2.2
TFD Arg
12.99
12.99
DL-Methionine
1.86
TFD Val
9.13
9.13
L-Lysine HCL
1.31
Calcium
7.6
7.6
L-Threonine
0.21
0.21
Avail. Phosphorus
3.8
3.8
DIAMOL
5.00
2.94
* 0.0 g/kg, 0.46 g/kg, 0.93 g/kg, 1.39 g/kg, and 1.85 g/kg DL-Methionine were added to the low
MC:L diet to achieve MC:L ratios of 63, 68, 73, 78, and 82%, respectively, in the experimental diets
III. RESULTS AND DISCUSSION
All performance criteria improved significantly and nonlinearly with increasing dietary
DSAA:DL (Table 2 and Figure 1). In most cases the treatment with extra Lys showed a better
performance providing evidence that Lys was the second limiting amino acid. Thus, it can be
concluded that the determined optimal ratios between true fecal digestible (TFD) sulphur
amino acids and TFD lysine is not underestimated in this trial.
Looking at the exponential regressions for live weight gain, 95% of asymptotic
response was reached at a DSAA:DL ratio of 75%. For FCR, carcass yield, and breast meat
yield, DSAA:DL ratio required for achieving 95% of asymptotic response was higher than
the higest tested ratio (82%) suggesting higher ratios to be optimal for these parameters in
male Ross 308 broilers. This high optimum is in agreement with the results reported by
Vieira et al. (2004) for both Ross 308 and Cobb 500 birds and Rostagno et al. (2007). In the
study reported by Lumpkins et al. (2007) where they studied the optimal level for digestible
sulphur amino acids, the ratio between total Methionine+Cysteine and Lysine at
recommended level for sulphur amino acids would suggest 86%.
91
Aust. Poult. Sci. Symp. 2010...21
Table 2.
Live weight gain, feed conversion ratio (FCR), carcass yield, and breast meat
yield in 35 days old male Ross 308 broilers fed diets with increasing levels of
true fecal digestible (TFD) sulphur amino acids (DSAA) to TFD lysine (DL)
ratios
DSAA : DL ratio
63%
68%
73%
78%
TFD Lysine, g/kg
9.44
9.44
9.44
9.44
TFD Met+Cys, g/kg
5.94
6.40
6.86
7.32
c
b
a
Gain, g
1489
1693
1778
1763a
c
ab
a
Feed intake, g
2723
2888
2981
2874b
FCR, kg/kg
1.813a
1.706b
1.677b
1.631c
c
b
b
Carcass yield, % of LW
69.75
70.65
71.11
71.23b
d
cd
bc
Breast meat, % of CW
30.66
31.65
32.39
33.02b
a
b
ab
Abdominal fat, % of CW
2.42
1.99
2.15
2.12b
Figures in rows with different superscript a, b, c differs significantly (P<0.05)
Live weight gain (d 14-35)
Live weight gain (g)
1800
1700
1600
1500
1400
1300
63
65
67
69
71
73
75
77
79
81
83
1.8
1.7
1.6
1.5
85
63
65
67
69
M+C-to-Lys ratio (%)
73
75
77
79
81
83
85
79
81
83
85
Breast Meat Yield (d 35)
34
Breast Meat Yield (% of CW)
Carcass Yield (% of LW)
71
M+C-to-Lys ratio (%)
Carcass Yield (d 35)
73
82%
9.44
7.78
1783a
2877b
1.614cd
71.54a
33.11b
1.95bc
Feed Conversion Ratio (d 14-35)
1.9
Feed Conversion Ratio (kg:kg)
1900
Extra Lys
74%
10.46
7.78
1819 a
2876 b
1.581 d
71.35 a
34.34 a
1.71 c
72
71
70
69
68
33
32
31
30
29
63
65
67
69
71
73
75
77
79
81
83
85
63
M+C-to-Lys ratio (%)
Figure 1.
65
67
69
71
73
75
77
M+C-to-Lys ratio (%)
Live weight gain, feed conversion ratio (FCR), carcass yield, and breast meat
yield in 35 days old male Ross 308 broilers in response to increasing levels of
Met+Cys:Lys ratio
In the present trial only the effect in males were studied but Rostagno et al. (2007)
reported that for FCR the optimal DSAA:DL ratio is actually higher in females compared to
male Cobb 500 broilers. In addition, Lumkins et al. (2007) found no difference in sulphur
amino acid requirements for female and male Cobb 500 broilers when using growth and
breast meat yield as parameter. However, they found higher sulphur amino acid requirements
for males when looking at FCR.
92
Aust. Poult. Sci. Symp. 2010...21
IV. CONCLUSION
The present results indicate that Ross 308 broilers in the grower period respond positively to
DSAA:DL ratios higher than currently recommended and that optimum varies with
performance parameter. For weight gain the optimal DSAA:DL ratio was found to be 75%
whereas for FCR, carcass yield and breast meat yield the optimum was even higher than the
highest ratio tested in this trial, i.e. 82%. Based on these data it seems like the economic
optimum will require a higher MC:L ratio especially if the production goal is processed
chicken compared to whole birds.
REFERENCES
Lumpkins BS, Batal AB, Baker DH (2007) Poultry Science 86, 325-330.
National Research Council (1994). Nutrient Requirements of Poultry. 9th ed. National
Academy Press, Washington, DC
Mack S, Bercovici D, DeGroote GE, Leclercq B, Lippens M, Pack M, Schutte JS, Van
Gauwenberghe S (1999) British Poultry Science 40, 257-265.
Rostagno H, Paez L, Albino L (2007) Proceedings, European Symposium on Poultry
Nutrition 16, 91-98.
Taherkhani R, Shivazad M, Zaghari M, Shahneh AZ (2008) The Journal of Poultry Science
45, 15-19.
Vieira SL, Lemme A, Goldenberg DB, Brugalli I (2004) Poultry Science 83, 1307-1313.
93
Aust. Poult. Sci. Symp. 2010...21
DIETARY ENZYMES ALTER SORGHUM PROTEIN DIGESTIBILITY AND AME
CONTENT
A. SULTAN 1, X. LI1, D. ZHANG1, D.J. CADOGAN 2 and W.L. BRYDEN1
Composition and availability of nutrients in sorghum are variable, especially protein content and
digestibility. Broilers fed sorghum based diets may under-perform, possibly reflecting low
nutrient availability due to the presence in sorghum of antinutritional factors; polyphenols,
phytate and kafirin (Bryden et al., 2009). The objective of this study was to assess the efficacy of
different dietary enzymes on apparent ileal protein digestibility (IPD) and metabolisable energy
(AME) content of sorghum. Eight mash diets were prepared with sorghum (918g sorghum/kg
diet) as the sole protein source. Celite (20 g/kg) was added to allow acid insoluble ash (AIA) to
be used as an indigestible marker. The treatments consisted of a control diet to which different
enzymes were added (see Table 1). Broilers, 36-days-old, were housed (7 birds/cage) in an
environmentally controlled shed and randomly assigned to replicated (n = 4) dietary treatments
with free access to feed and water. On day-42, birds were euthanized and contents of the lower
half of the ileum were pooled per pen, frozen and lyophilized. Nitrogen, gross energy and AIA
content of all diet, ileal and faecal samples were determined using standard laboratory protocols
and protein digestibility coefficients and AME values were calculated (Table 1).
Table 1.
Effect of dietary enzymes on AME and apparent ileal protein digestibility of
sorghum in broilers.
Treatment
Ileal protein digestibility
AME
coefficient
(MJ/kg DM)
1 Control
0.780bc
14.07e
c
2 Xylanase
0.772
14.37d
ab
3 Phytase
0.808
14.62c
a
4 Protease
0.815
14.81bc
abc
5 Xylanase + Phytase
0.805
14.75c
abc
6 Xylanase + Protease
0.797
14.66c
7 Phytase + Protease
0.825a
14.99ab
ab
8 Xylanase + Phytase + Protease
0.813
15.18a
SEM
0.01
0.06
abcde
Mean values within columns not sharing superscripts are significantly different (P < 0.05)
The apparent ileal protein digestibility coefficient (IPD) was significantly enhanced by
5.47% and 4.33% in treatments 7 and 4, respectively. The presence in the diet of all three
enzymes resulted in an improvement of 4.02% in the IPD coefficient. All enzymes singly and in
combination significantly improved AME. Maximum improvement (7.89%) in AME was seen in
treatment 8, with all three enzymes in combination followed by treatments 7 (6.54%) and 4
(5.26%), respectively. These findings demonstrate that strategic application of enzymes to
sorghum based poultry diet can reduce the negative influence of antinutritional factors, thus
enhancing nutrient digestibility and bird performance.
Bryden WL, Selle PH, Cadogan DJ, Li X, Muller ND, Jordon DR, Gidley MJ Hamilton WD
(2009) RIRDC Publication 09/077. Barton, ACT.
1
2
The University of Queensland, Gatton, QLD 4343
Feedworks, Romsey Vic 3434.
94
Aust. Poult. Sci. Symp. 2010...21
INFLUENCE OF ENERGY AND BALANCED PROTEIN LEVELS IN WHEAT-BASED
DIETS ON ROSS 308 BROILER PERFORMANCE
T.G. MADSEN 1, S. CARROLL 2, C. KEMP2 and A. LEMME 3
Summary
An experiment with male and female Ross 308 broilers was performed to investigate the
interactions between graded levels of balanced protein (BP) and metabolizable energy (ME)
level. At 42 days of age feed intake, live weight and feed conversion ratio (FCR) were
significantly affected by BP, ME and sex of the birds (P<0.05). With increasing BP level feed
intake declined and live weight and FCR improved independently of the dietary ME level.
When ME was reduced the feed intake increased, FCR was impaired and live weight
remained relatively stable. Breast meat yield increased with increasing BP level whereas
increasing ME level had a negative effect on breast meat yield. The data do not indicate an
optimal ME to BP ratio and economic optimum for both energy and balanced protein will
therefore depend on the price ratio of protein and energy in feed ingredients.
I. INTRODUCTION
Prices for cereals and oils have fluctuated considerably over the last few years and have in
general led to higher unit costs for the energy components of broiler diets. Thus, a reduction
of the energy level of broiler diets seems attractive in order to control feed costs. It is
commonly recognized that broilers adjust their feed intake in accordance with the dietary ME
content; complete adjustment would keep ME intake constant (Leeson et al. 1996). Based on
such findings it has been suggested that, as ME in varied, BP should be kept in a constant
ratio to ME. However, Plumstead et al. (2007) found a positive effect of increasing BP which
was independent of ME level and in this study they did not see an effect of dietary ME level
on feed intake. Kamran et al. (2008) found that broilers given feed with constant ME:CP
ratios were not able to fully compensate for a severe decrease in dietary ME level by
increasing feed intake. Thus, it still remains unclear how BP should be adjusted when ME
level is changed and therefore a trial was conducted to test the effect of graded levels of BP
fed at two dietary ME levels on weight gain, feed conversion ratio, and carcass characteristics
in broilers.
II. MATERIAL AND METHODS
A total of 4320 male and 4320 female day old Ross 308 broiler chicks (44 g) were housed in
96 floor pens (2.7 x 2.1 m; 90 chicks per pen) at the research facility of AVIAGEN Ltd.,
Newbridge, UK. The chicks received a common wheat based starter diet (sieved crumble, d
1 to 10). From d 11 a 2 x 4 factorial approach with 2 energy levels (95 and 100 % of Aviagen
recommendation) and 4 levels of balanced protein (80, 90, 100, and 115 % of Aviagen
recommendation) was applied resulting in 8 dietary treatments per sex and 6 replicate pens
per treatment. Environmental conditions (temperature, humidity, light program) were in line
with the recommendations given in the Ross Management Guide (Aviagen, 2002).
1
Health and Nutrition, Evonik Degussa (SEA) Pte. Ltd, Singapore
Aviagen Ltd., Scotland
3
Health and Nutrition, Evonik Degussa GmbH, Hanau, Germany
2
95
Aust. Poult. Sci. Symp. 2010...21
The grower diets were fed from day 11 to 25 and finisher diets were fed from day 26
to 42. The pelleted grower and finisher feeds as well as water were offered ad libitum. Pellet
quality was good due to the high inclusion levels of wheat. However, in the grower diets
sieve analysis revealed a less stable pellet with increasing dietary fat content (increasing ME).
This effect was not observed in the finisher diets.
Diets were mainly based on wheat and soybean meal and the ME level was reduced
by decreasing fat and increasing wheatfeed (Table 1). At both ME levels the BP content was
stepwise increased by increasing the levels of soybean meal, maize gluten meal and fish
meal. The inclusion level of fat had to be increased at the high BP diets due to lower ME
content of soybean meal compared to wheat.
Performance data were subjected to ANOVA analyses using Genstat (9th Edition)
using the following statistical model:
Yijkl = μ + Ri + Bj + Fk + R*Bij + R*Fik + B*Fjk + R*B*Fijk + eijkl ,
where Yijkl is a specific trait per experimental unit (pen of birds), μ the overall mean, Ri is the
ME effect (i = 95 or 100), Bj is the effect of BP level (j = 80, 90, 100 and 115), Fk is the effect
of sex (k = female or male), R*Bij, R*Fik, B*Fjk, and R*B*Fijk the interactions between the
factors, and eijkl is the residual error term.
Table 1.
Ingredient and nutrient composition of experimental diets with lowest and
highest balanced protein (BP) levels and the two metabolizable energy (ME)
levels
Grower (d 11 – 25)
Finisher (d. 26 – 42)
ME
100 %
95 %
100 %
95 %
BP1
80% 115%
80%
115%
80%
115%
80% 115%
Diet composition, g/kg
Wheat
696
470
659
437
731
528
695
497
Soybean meal
183
365
177
355
142
313
141
303
Wheat feed
20.0
20.0
80.0
80.0
20.0
20.0
80.0
80.0
Corn gluten meal
5.0
10.0
5.0
10.0
5.0
10.0
5.0
10.0
Fish meal
5.0
15.0
5.0
15.0
5.0
10.0
5.0
10.0
Vit. and Min.
39.2
35.6
38.9
35.3
37.6
35.1
37.3
34.8
Soya oil
20.0
20.0
10.0
10.0
20.0
20.0
10.0
10.0
Vegetable fat
25.7
57.2
18.9
49.8
28.3
57.5
21.0
49.4
Liq. L-Lys (50%)
3.5
2.6
3.5
2.7
3.3
2.4
3.2
2.5
DL-Methionine
1.8
3.5
1.8
3.4
1.5
3.1
1.5
3.1
L-Threonine
0.8
0.9
0.8
1.0
0.7
0.9
0.7
0.9
Calculated nutrient contents, g/kg
Crude Protein
168
247
171
248
153
223
156
224
ME, MJ/kg
13.19
13.19
12.53
12.53
13.40
13.40
12.73
12.73
TFD2 Lys
8.8
12.7
8.8
12.7
7.8
11.2
7.8
11.2
TFD Met+Cys
6.7
9.6
6.7
9.6
6.1
8.7
6.1
8.7
TFD Thr
5.8
8.3
5.8
8.3
5.2
7.5
5.2
7.5
1
Energy and balanced protein target levels are given as % of Aviagen recommendation (Aviagen,
2002). 2True Fecal Digestible
1
III. RESULTS AND DISCUSSION
Reducing the dietary ME level from 100 to 95 % significantly (p < 0.001) increased feed
intake at all dietary protein levels. However, the increased feed intake did not entirely
compensate for the reduced dietary ME level and thus the ME intake decreased on average by
2.2 %. This response is in contrast to former experiments where broilers fully compensated
for a 5% reduction in dietary energy content in pelleted diets (Lemme 2004; Huang et al.
96
Aust. Poult. Sci. Symp. 2010...21
2008). At both energy levels, the feed intake decreased as the dietary content of BP increased
and consequently the energy intake decreased as well. In contrast, the protein intake
increased linearly as the reduced feed intake was more than compensated by the increased
dietary protein content. Live weight increased significantly as the BP supply increased (p <
0.05), however differences between the two ME levels became smaller as the BP level
increased (see Figure 1). Although 15 % above the Aviagen recommendations the highest
level of BP led to a numeric increase in live weight gain at both ME levels. The FCR was
impaired as the dietary ME level was reduced (p < 0.01). On the other hand, increasing BP
supply improved FCR (p < 0.05) which is in line with previous trial results (Lemme et al.
2006) and the findings of Plumstead et al. (2007). Even the highest BP level caused a further
improvement of the FCR although it was above the recommended level.
Table 2.
M
Energy1
(%)
95
M
100
F
95
F
100
Sex
P values
1
Effect of energy and balanced protein (BP) level on live weight (LW), feed
conversion ratio (FCR), carcass yield and breast meat yield in female (F) and
male (M) Ross 308 broilers at day 42
BP1
(%)
80
90
100
115
80
90
100
115
80
90
100
115
80
90
100
115
LW
(g)
3159
3357
3346
3433
3301
3330
3397
3446
2696
2745
2802
2774
2727
2759
2761
2805
FCR2
(g)
1.89
1.82
1.76
1.69
1.79
1.75
1.70
1.63
1.90
1.84
1.80
1.76
1.83
1.79
1.77
1.72
Carcass
(% of LW)
65.8
65.9
66.0
66.7
65.9
65.9
66.4
67.0
66.6
67.2
67.4
67.5
66.5
66.8
67.3
67.2
Breast
(% Carcass)
24.9
25.5
26.7
27.3
24.6
25.1
26.0
27.3
25.5
26.6
27.3
27.8
25.4
25.9
27.6
27.9
ME
BP
Sex
ME*BP
ME*Sex
BP*Sex
ME*BP*Sex
0.006
0.001
0.001
0.004
0.058
0.001
0.010
0.001
0.001
0.001
0.157
0.015
0.010
0.756
0.947
0.001
0.001
0.809
0.202
0.402
0.976
0.044
0.001
0.001
0.362
0.334
0.302
0.299
ME and BP levels are given in % of Aviagen recommendation. 2 corrected for mortality
The dietary level of BP significantly increased carcass yield (p < 0.001) whereas the
ME level had no effect (p > 0.05). Furthermore, reducing the ME level as well as increasing
the dietary BP supply decreased the abdominal fat and increased breast meat significantly (p
< 0.05). This is in line with the findings of Lemme (2004) and suggests that the genetic
97
Aust. Poult. Sci. Symp. 2010...21
potential of the broilers was not fully expressed by feeding the 100 % BP diet.
2.0
Feed conversion Ratio (g:g at 42d)
Live weight (g at 42d)
3500
3300
Male_95
Male_100
Female_95
Female_100
3100
2900
2700
1.8
1.7
1.6
1.5
2500
70
80
90
100
110
Balanced Protein (% of Aviagen rec)
70
120
20.0
80
90
100
110
Balanced Protein (% of Aviagen rec)
120
68.0
19.0
Carcass yield (% of LW)
Breast meat yield (% of carcass)
Male_95
Male_100
Female_95
Female_100
1.9
18.0
17.0
Male_95
Male_100
Female_95
Female_100
16.0
15.0
70
Figure 1.
80
90
100
110
Balanced Protein (% of Aviagen rec)
67.0
66.0
Male_95
Male_100
Female_95
Female_100
65.0
64.0
120
70
80
90
100
110
Balanced Protein (% of Aviagen rec)
120
Live weight (top left), mortality corrected feed conversion ratio (top right),
breast meat yield (bottom left), and carcass yield (bottom right) in 42 days old
female (■, □) and male (●, ○) Ross 308 broilers in response to increasing
levels of balanced protein at two energy levels (95% (■, ●) and 100% (□, ○))
IV. CONCLUSION
The present data show that broilers did not completely compensate ME intake by increased
feed intake when dietary content of ME was reduced by 5% compared to recommended
levels. Moreover, increasing dietary BP improved weight gain and particularly feed
conversion ratio independently from the dietary ME level. Thus, economic optimum for both
ME and BP will depend on the price ratio of protein and ME in feed ingredients and one
optimum can not be given.
REFERENCES
Aviagen (2002) Ross Broiler Management Manual, AviagenLtd., Newbridge, UK
Huang KH, Fleming E, Kenny M (2008) Proceedings, 23rd World Poultry Conference
Kamran Z, Sarwar M, Nisa M, Nadeem MA, Ahmad S, Mushtaq T, Ahmad T, Shahzad MA
(2008) Asian-Australian Journal of Animal Science 21, 1629-1634.
Leeson S, Caston L, Summers JD (1996) Poultry Science 75, 529-535.
Lemme A (2004) Facts and Figures 1544, www.aminoacidsandmore.com Evonik Degussa
GmbH, Hanau, Germany
Lemme A, Wijtten PJA, van Wichen J, Petri A, Langhout DJ (2006) Poultry Science 85, 721730.
Plumstead PW, Romero-Sanchez H, Paton ND, Spears JW, Brake J (2007) Poultry Science
86, 2639-2648.
98
Aust. Poult. Sci. Symp. 2010...21
VARIATION IN NUTRIENT COMPOSITION AND STRUCTURE OF HIGH-MOISTURE
MAIZE DRIED AT DIFFERENT TEMPERATURES
M.M. BHUIYAN 1, P.A. IJI1, A.F. ISLAM1 and L.L. MIKKELSEN1
Summary
The chemical composition and structure of high moisture maize grains were investigated
after drying at different temperatures and compared to sun-dried maize. The chemical
composition was affected by artificial drying of high-moisture maize grain. It is thought that
this may have some ramification for the nutritive value of the grain when fed to chickens.
I.INTRODUCTION
In many parts of the world maize (Zea mays) is harvested at relatively high moisture (HM)
content, with a view to minimizing damage in the field when left to dry naturally. The grain
is then subjected to artificial drying, which may result in loss of quality such as increase in
retrograde starch content (Brown, 1996). Heat processing may also anneal the starch as the
grain cools down. The digestibility of cereal grains is influenced by the starch component,
especially the ratio between amylose and amylopectin (McDonald et al., 1995) Due to its
amorphous nature, amylopectin is more readily digested than the amylose. The normal
structure (spherical, 10-16 microns across) of starch granules with protein bodies and matrix
may be influenced easily and can create a favorable environment for enzymatic digestion
(Taylor and Belton, 2002). The present study was carried out with a view to investigate the
variation in chemical composition and structure of the HM maize grains and how this affects
on the nutritional value subsequently.
II. MATERIALS AND METHODS
Maize grain, from the 2009 planting year, was obtained from Inverrel in northern NSW at
around 23% moisture. After collection, maize cobs were split into four groups and dried
under the sun or artificially in an oven at 80, 90 or 100oC for 24 hours. The proximate
nutrient composition of the maize samples were determined using standard methods (AOAC,
2002). The gross energy (GE) contents of the samples were determined using an IKA® WERKE bomb calorimeter at UNE. Metabolizable energy (ME) was calculated according to
the equation [ME (Kcal/kg) = 53+38 × (%CP +2.25 × %EE +1.1×%Starch +%Sugar)]
developed by Carpenter and Clegg (1956).
Total and resistant starch contents were determined according to the method of
McCleary et al. (1994). Amylose/amylopectin ratio was also determined with Megazyme
amylose/amylopectin assay kit using the selective quantitative precipitation reaction of
concanavalin A (ConA) with amylopectin according to Gibson et al. (1996).
The concentrations of amino acids were determined at the Australian Proteome
Analysis facility at Macquarie University using pre-column derivatisation of amino acid with
6-aminoquinoly-N-hydroxysuccinimidyl carbamate (AQC) followed by separation of the
derivatives and quantification by reversed phase high performance liquid chromatography.
Mineral elements were analyzed by inductively coupled plasma (ICP) method (Vista MPXradial) and the sealed digest chamber digest (SCD) at UNE on the basis of Anderson and
Henderson (1986).
1
School of Environmental and Rural Sciences, University of New England, Armidale 2351 NSW
99
Aust. Poult. Sci. Symp. 2010...21
Non-starch polysaccharide (NSP) contents were measured by gas chromatography
(VARIAN, CP-3800, USA) at UNE as outlined by Englyst and Hudson (1993). The in vitro
digestibility of dry matter and starch was determined by the method of Babinszky et al.
(1990) and viscosity according to Bedford and Classen (1993). The grains were scanned on a
NeoScope, JCM-5000 table-top SEM (JEOL Ltd, Tokyo, Japan). Whole grain samples were
scoured around the edges and cut in sections, then mounted on the machine for assessment
with a magnification of x1000. All the collected data were subjected to non-parametric
analyses using SPSS (17th Version) followed by calculations of coefficient of variation (CV).
III. RESULTS AND DISCUSSION
The proximate composition of the different batches of maize is shown in Table 1. The DM
and ash contents were increased by artificial drying of the grains at all temperature,
compared to sun drying. However, the CP, CF and phytate contents were decreased by up to
4.7, 8.2 and 32.8%, respectively, as a result of artificial drying of the grains. The GE content
was decreased by 2.2% but ME content, estimated from nutrient composition, was increased
by 2.1%. Artificial drying also increased the concentrations of most amino acids.
Table1.
Chemical composition of maize batches observed under different drying
temperatures
A. Dry matter, crude protein, crude fat, ash, phytate-P (g/kg DM) GE and ME (MJ/kg DM)
DM
CP
CF
Ash
Phytate-P
GE
ME1
Sun-dried
870.0
98.4
45.0
1.23
1.80
18.86
15.42
o
80 C
950.0
93.4
42.0
1.29
1.36
18.36
15.58
90 oC
963.0
92.2
42.1
1.30
1.35
18.41
15.50
o
100 C
980.0
93.8
41.3
1.32
1.21
18.45
15.74
CV
0.05
0.03
0.04
0.03
0.18
0.01
0.01
B. Amino acids (g/kg DM)
Met
Lys
Thr
Ala
Phe
Arg
Leu
Isoleu
Val
Sun-dried
1.2
2.2
3.0
6.4
4.3
4.2
10.3
3.3
4.1
80 oC
1.4
2.0
3.1
6.8
4.4
4.3
11.2
3.7
4.3
90 oC
1.4
1.9
3.2
6.9
4.6
4.2
11.7
3.6
4.4
100 oC
1.3
1.9
3.3
7.0
4.7
4.3
11.9
3.8
4.4
CV
0.08
0.08
0.03
0.04
0.04
0.01
0.07
0.06
0.03
C. Starch content and components (g/kg DM)
Starch
Resistant
Amylopectin
Amylose
Amylose:
starch
Amylopectin
Sun-dried
670.1
316.6
506.7
280.3
0.55
80 oC
691.3
362.7
424.1
303.7
0.72
90 oC
687.8
366.3
369.0
308.0
0.83
100 oC
684.0
415.7
306.6
313.8
1.02
CV
0.01
0.11
0.21
0.05
0.25
D. Minerals
Macro-elements (g/kg DM)
Micro-elements (mg/g DM)
Ca
P
Na
K
Mg
Mn
Zn
Fe
Cu
Mo
Sun-dried
0.05
2.9
7.9
3.5
1.6
6.2
17.0
37.4
1.0
0.43
80 oC
0.05
2.6
7.3
3.3
1.4
8.7
15.4
27.9
1.1
0.43
o
90 C
0.04
2.5
7.1
3.1
1.3
5.1
14.0
25.2
1.1
0.39
100 oC
0.03
2.6
7.0
3.2
1.4
5.6
14.9
18.7
1.5
0.39
CV
0.23
0.08
0.05 0.05
0.08
0.25
0.08
0.28
0.19
0.06
1
CV = Coefficient of variation, According to chemical analysis (Carpenter and Clegg, 1956).
100
Aust. Poult. Sci. Symp. 2010...21
Total starch, resistant starch and amylose contents were increased by 2.1, 31.3 and
12.0%, respectively for the grain dried at 100 oC, compared to the sun-dried grain.
Conversely, the content of amylopectin was decreased by 39.5%.
Mineral content was adversely affected by artificial drying of the grain. For example,
the concentrations of Ca, P, Na, K and Mg were reduced by 40.0, 10.3, 11.4, 8.6 and 12.5%,
respectively as a result of drying at 100 oC, when compared to sun drying.
Total soluble NSP content varied from 3.29 to 3.79 g/kg, being generally higher in the
artificially dried grains than in the sun-dried samples (Table 2). Ribose and arabinose were
the most variable of the component sugars, in terms of the CV. There was very little
variation in the concentrations of insoluble NSP, with CV generally lower than 0.05.
Table 2.
Non- Starch Polysaccharide (NSP) components of maize batches observed
under varying drying temperatures
A. Soluble NSP content and components (g/kg DM)
Rib
Ara
Xyl
Sun-dried
0.05
0.55
0.35
80 oC
0.08
0.77
0.46
90 oC
0.07
0.76
0.46
o
100 C
0.08
0.85
0.49
CV
0.19
0.17
0.14
B. Insoluble NSP content and components (g/kg DM)
Ara
Xyl
Man
Sun-dried
20.08
25.34
1.00
80 oC
19.42
24.68
0.94
90 oC
19.56
25.27
0.93
100 oC
19.38
25.29
0.93
CV
0.02
0.01
0.04
Man
1.89
1.84
1.86
1.83
0.05
Gal
0.34
0.43
0.40
0.39
0.10
Gal
5.71
5.29
5.50
5.38
0.03
Glu
0.50
0.60
0.58
0.61
0.09
Glu
22.34
21.98
22.93
22.60
0.02
NSP
3.29
3.87
3.69
3.79
0.07
NSP
66.11
64.21
65.87
65.33
0.01
In vitro digestibility indicated (Table 3) that digestibility of DM was improved by
artificially drying the HM maize but the starch digestibility was reduced (Table 3). There
was no effect of treatment on in vitro digesta viscosity.
Table 3.
In vitro digestibility of dry matter and starch as well as viscosity of maize
batches observed under varying drying temperatures
Viscocity (cp)1
Digestibility
Treatment
Dry matter
Starch
Sun-dried
48.0
80 oC
52.4
90 oC
52.3
o
100 C
52.5
CV
0.04
1
cp (centipoise) = 1/100 dyne second per centimeter2
63.0
58.2
57.6
58.4
0.04
0.92
0.91
0.91
0.90
0.009
The scanning micrographs of the different grains show variations in the morphology
of the starch granules (Plate 1). The granules shrunk in size as a result of artificial drying of
the grains. The internal matrix of the grains, which may reflect the linkage of starch and
protein bodies, was also different.
101
Aust. Poult. Sci. Symp. 2010...21
Plate 1.
Electronic microscope images (x1000) of sun-dried (top left) or grain dried at
80oC (top right), 90oC (bottom left) and 100oC (bottom right).
IV.CONCLUSIONS
It can be concluded that artificial drying of HM maize alters the composition of the grain.
Further study can be conducted to ascertain if the nutritive value of the grain is affected, or if
it could be improved through nutritional intervention.
REFERENCES
Anderson DL, Henderson LJ (1986) Agronomy Journal 78, 937-938.
AOAC (2002) Official Methods of Analysis of AOAC Inernational (17th edition).
Babinsky L, Van der Meer JM, Boer H, Hartog LA (1990) Journal of the Science of Foood
and Agriculture 50, 173-178.
Bedford MR, Classen HL (1993) Poultry Science 72, 137-143.
Brown I (1996) Nutrition Reviews 54, S115-S119.
Carpenter KJ, Clegg KM (1956) Journal of the Science of Foood and Agriculture 7, 45.
Englyst HN, Hudson GJ (1993) In Dietary Fibre and Human Nutrition [GA Spiller, Ed.]
Gibson TS, Solah VA, McCleary BV (1996) Journal of Cereal Science 25, 111-119.
McCleary, B. V., Solah, V., & Gibson, T. S. (1994) Journal of Cereal Science 20, 51-58.
McDonald P, Edwards RA, Greenhalgh JFD, Morgan CA (1995) In: Animal Nutrition.
Longman Scientific and Technical. Harlow, UK.
Taylor JRN, Belton PS (2002) In: Pseudocereals and Less Common Cereals. [PS Belton, JRN
Taylor Eds.] pp 25-91. Springer, Berlin,
102
Aust. Poult. Sci. Symp. 2010...21
INFLUENCE OF FEED FORM ON INTAKE PREFERENCE AND PERFORMANCE OF
YOUNG BROILERS
K.H. HUANG 1 and M. DE BEER1
Summary
A series of experiments was conducted to investigate feed particle size preference and the
effect of feed form on performance of young broilers. Feed particle size significantly
influenced feed intake preference of modern young broilers, and affected their ability to
achieve genetic potential weight gain at both 7days and at market age. In experiment 1 with
broilers aged between 1 and 14 days, the birds preferred a particle size between 0.86mm and
2.00mm. As birds grew, they tended to prefer bigger particle sizes. Birds did not select fine
particles (< 0.86 mm) even at 3 days of age. Experiment 2 studied the performance of broilers
between 0 and 9 days of age. Feeds of different particle sizes significantly (P < 0.05)
influenced performance over this period. In experiment 3, comparing crumble feed with cold
mash, the crumble significantly increased body weight and feed intake, while feed conversion
was significantly better in crumble, compared coarse mash, but similar to find mash. It was
concluded that young broilers have preferences for feed particles of a certain size to
maximize feed intake and subsequently this behaviour influenced their growth performance.
With broilers reaching market weight at younger ages, the improved early growth as affected
by particle size and feed form would be of commercial significance.
I. INTRODUCTION
Feed physical quality has a major influence on modern broiler performance. Published
literature shows that broilers fed with good-quality pellets have better growth performance
and feed conversion than those fed with mash, reground pellets, or pellets with more fines
(Huang et al., 2008; Kenny and Flemming, 2006; Lemme et al., 2006). However, there is less
clarity about how feed particle size influences broiler feeding behaviour and about how
responses to feed physical quality in the early stages of growth, up to 14 days, are reflected in
performance at later ages. In general 7-day body weight is strongly correlated with body
weight at market age. The work reported here investigated the effect of feed form on young
broiler feed choice, and on biological performance.
II. FEED PREPARATION
Corn-soybean meal based broiler starter and grower feeds, formulated to Aviagen Broiler
Nutrition Specifications (2007), were pelleted (3 mm die) and then crumbled. The crumbled
feed was then passed through a series of sieves in experiment 1 and 2 in order to separate
different size fractions. The feeds for experiment 1 and 2 were mixed to the same
specifications and separated by the same methods. After sieving there were four feed
fractions in experiment 1 and three fractions in experiment 2 as shown in Table 1. In
experiment 3, the particle size analysis of crumble and mash feeds was as shown in Table 2.
In experiment 1 and 2, the different fractions were tested for moisture, crude protein, fat,
fibre, calcium and phosphorous. In experiment 3, both feeds were tested for moisture, crude
protein, fat, calcium and sodium. There were no significant differences in the contents of any
major nutrients between the separated fractions or feed forms (data not shown).
1
Aviagen Inc., Cumming Research Park, 5015 Bradford Drive, Huntsville, AL. USA
103
Aust. Poult. Sci. Symp. 2010...21
Table 1.
The fractions of feed particle size for Experiment 1 and 2.
Experiment
Feed particle size (mm)
Starter (1-10 days)
Grower (11-14 days)
Experiment 1
< 0.86
0.86 - 2.00
2.00 - 3.18
> 3.18
Experiment 2
< 0.86
0.86 – 2.00
2.00 – 3.18
Table 2.
The distribution of feed particle size in Experiment 3 (starter 1-10days).
Feed particle size
(mm)
< 1.00
1.00 – 2.00
2.00 – 3.00
> 3.00
< 2.00
2.00 – 3.18
3.18 – 4.76
> 4.76
Crumble
(%)
Fine mash
(%)
Coarse mash
(%)
22.73
32.25
21.18
24.84
96.00
4.00
0.00
0.00
41.44
24.74
17.36
16.46
III. FEED PARTICILE SIZE PREFREENCE OF YOUNG BROILERS
Experiment 1, utilized a total of six pens with 50 straight run birds per pen. All pens had
identical nipple drinkers and brood lights. Each pen contained four feeders with one each of
the four different size fractions of starter or grower feeds. All feed fractions and water were
offered ad libitum. The feeders were placed equidistant from the brood light and randomly
rotated so that no feeder was in the same location for two days running. This prevented
chicks from “learning” where their preferred particle size was located. The results are shown
in Table 3, expressed as relative percentage of total intake for each period. At each age there
were significant differences in the amount of each fraction consumed. and, as birds got older,
they increased consumption of bigger particle sizes. Birds were eating significantly more feed
of particle size between 0.86 to 3.18 mm for each age studied.
IV. FEED FORM AND BROILER PERFORMANCE
In experiment 2, a total of 12 pens with 50 straight run birds per pen were utilized. Three
separated feed fractions were compared (Table 1), with all material over 3.18 mm being
discarded. Each fraction was fed separately to 4 replicate pens. All feed fractions and water
were offered ad libitum. Feed intake, body weight and feed conversion ratio (FCR) were
determined for each pen at 5 and 9 days of age. The results were shown in Table 4. Feed
particle size significantly (P < 0.05) affected live weight, feed intake and FCR at 9 days, but
not at 5 days (P > 0.05). The particle size between 2-3.18 mm achieved the best body weight
and FCR among the treatments, while particle size less than 0.86 mm gave the poorest
performance.
104
Aust. Poult. Sci. Symp. 2010...21
Table 3.
Feed particle
size (mm)
< 0.86
0.86 - 2.00
2.00 - 3.18
> 3.18
SEM
abc
Feed particle size preference of broiler chicks between days 1 to 14
(Experiment 1).
25.19b
38.90a
21.60c
14.32d
1.634
21.87b
49.57a
16.82bc
11.74c
1.813
9.68c
53.97a
20.59b
15.76bc
2.301
Feed intake day
11-14 (%)
< 2.00
2.00 - 3.18
3.18 - 4.76
> 4.76
SEM
22.02c
42.12a
33.91b
1.95d
1.95
P
0.001
P
0.001
0.001
0.001
values without common superscript differ significantly (P < 0.05).
Table 4.
Grower (day 11-14)
Starter (day 1-10)
Feed particle
Feed intake Feed intake day Feed intake day size (mm)
day1-3 (%)
4-6 (%)
7-10 (%)
Effect of feed particle size on broiler performance at 5-days and 9-days
(Experiment 2)
5 Days
Live
Feed
Live
Feed particle
Weight
Intake
FCR
Livability weight
size
(g)
(g)
(g/g)
(%)
(g)
2.0 - 3.18mm
112.5
62.55
0.555
99.5
233.8a
0.86 - 2.0mm
108.5
61.675
0.566
99
215.0b
< 0.86mm
104.75
61
0.583
99.5
194.0c
SEM
2.727
0.859
0.0172
0.726
2.827
P
0.213
0.485
0.543
0.857
0.001
abc
values without common superscript differ significantly (P < 0.05).
9 Days
Feed
Intake
(g)
140.3a
132.9b
126.6c
1.683
0.004
FCR
(g/g)
0.869c
0.902b
0.967a
0.00916
0.001
Livability
(%)
99.0
98.0
99.5
0.500
0.178
V. FEED FORM AND BROILER FEEDING BEHAVIOR
Experiment 3, utilized a total of 24 pens with 16 male birds per pen to give 8 replicates per
treatment in a randomised block design. Three treatments consisted of crumble, coarse and
fine mash. The particle size distribution is shown in Table 2. All feeds and water were offered
ad libitum. Feed intake, body weight, FCR, and liveability were determined for each pen at
10 days of age. The results are presented in Table 5. Live weight, feed intake and FCR were
significantly (P < 0.05) affected by feed form, with best performance on the crumble
treatment. FCR was significantly (P < 0.05) better in crumble, compared coarse mash, but
similar (P > 0.05) to fine mash.
105
Aust. Poult. Sci. Symp. 2010...21
Table 5.
Effect of feed form on broiler performance at 10days (Experiment 3)
Live Weight
(g)
Feed Intake
(g)
FCR
(g/g)
Liveability
(%)
327.5a
289.0b
289.6b
2.981
300.4a
284.1b
264.4c
2.785
1.064b
1.165a
1.079b
0.0105
100
99.22
100
0.451
P
0.001
0.001
0.001
values without common superscript differ significantly (P < 0.05).
0.393
Treatment
Crumble
Coarse Mash
Fine Mash
SEM
abc
VI. CONCLUSION
The present results demonstrate that young broilers do show preference for feed particle size
and this preference tends to change to bigger particle size as the birds age. Birds rejected fine
crumbles (< 0.86 mm) even at 3 days of age. When the birds were provided with the
preferred crumble size, their performance was significantly (P < 0.05) improved at 9 days.
Feed form is important for modern broilers to achieve maximum growth, even at a very
young age.
REFERENCES
Ross Broiler Nutrition Specification (2007) Aviagen Inc., Huntsville, Alabama, USA.
Huang KH, Fleming E, Kenny M (2008) Proceedings, 23rd World’s Poultry Congress p. 250
(Abstr).
Kenny M, Flemming E (2006) Proceedings, Australian Poultry Science Symposium 18, 2529.
Lemme A, Wijtten PJA, van Wichen J, Petri A, Langhout DJ (2006) Poultry Science 85, 721730.
106
Aust. Poult. Sci. Symp. 2010...21
DIFFERENTIATION BETWEEN PATHOGENIC SEROTYPE 1 ISOLATES OF
MAREK’S DISEASE VIRUS (MDV1) AND THE RISPENS VACCINE IN AUSTRALIA
USING REAL-TIME PCR.
K.G. RENZ 1, B. F. CHEETHAM 2, and S. W. WALKDEN-BROWN1
Summary
Two real-time PCR assays were developed which enable quantification and differentiation
between pathogenic Australian isolates of MDV serotype 1 and the serotype 1 vaccine strain
Rispens/CVI988. The assays are based on a DNA sequence variation in the meq gene
between pathogenic and vaccinal MDV1 which has been confirmed by sequencing of 20
Australian field strains of MDV. Complete specificity has been demonstrated in samples
containing pathogenic MDV (n=20), Rispens (3 commercial vaccine strains), or both. The
limit of detection of both the Rispens-specific and the pathogenic MDV1-specific assays was
10 viral copies/reaction.
I. INTRODUCTION
Marek’s disease virus (MDV) is classified as an Alphaherpesvirus and isolates of MDV can
be classified into three serotypes, namely serotypes 1 (MDV1), 2 (MDV2) and 3 (herpesvirus
of turkeys or HVT) of which only MDV1 is pathogenic and oncogenic (Schat and Calnek,
1978). Since the detection of the aetiological agent in the late 1960s (Churchill and Biggs,
1967), MD has been controlled to a large extent using vaccines which consist either of
attenuated isolates of oncogenic MDVs or the apathogenic HVT or MDV2 serotypes (Witter
et al., 1970; Schat and Calnek, 1978).
Vaccination against Marek’s disease using live vaccines provides protection against
clinical MD but not against co-infection with wild-type pathogenic MDVs which continue to
multiply in the host and be shed in feather dander at very high levels (Islam and WalkdenBrown 2007). Thus vaccinated chickens may harbour mixed populations of MDVs with
considerable implications for diagnosis of infection. In Australia broiler chickens either
remain unvaccinated or are vaccinated in ovo with HVT while layers and broiler breeders
usually are vaccinated with the MDV1 Rispens/CVI988 vaccine for long-term protection.
Rispens/CVI988 is the dominant attenuated MDV1 vaccine in use worldwide.
Several molecular tests using the real-time quantitative PCR (qPCR) technique are
available in order to differentiate between the three MDV serotypes (eg. Islam et al., 2006,
Renz et al., 2006), but there is no quantitative test which will differentiate between
pathogenic MDV1 and Rispens/CVI988. Previous tests to differentiate Rispens from
pathogenic MDV were based on the number of 132 bp repeats (Becker et al. 1992). However
these assays were not able to quantify virus and it has subsequently been shown that this
marker is lost in as little as one back passage in chickens (Young and Gravel 1996). There is
a preliminary report of a qPCR test to differentiate wild type MDV1 from Rispens vaccine
based on the pp38 and ICP4 genes (Zelnik et al., 2008) but it is unclear that assay
development is complete.
This paper describes the first molecular assay using the qPCR technique to
differentiate between Australian field isolates of MDV1 and the Rispens vaccine.
1
Animal Science, School of Environmental and Rural Science, University of New England, Armidale,
NSW 2351, Australia
2
Molecular and Cellular Biology, School of Science and Technology, University of New England, Armidale,
NSW 2351, Australia
107
Aust. Poult. Sci. Symp. 2010...21
II. MATERIAL AND METHODS
DNA from six Australian isolates of MDV1, namely MPF57, Woodlands1, MPF132/5,
02LAR, 04KAL, FT158 was obtained from spleen samples collected at 13 days post infection
(dpi) from experimentally infected unvaccinated specific pathogen free (SPF) chickens
involved in a pathotyping experiment (Walkden-Brown et al., 2006). The isolates induced
gross MD lesions in 52.9-94.4 % of unvaccinated chickens and the protective index provided
by HVT vaccination varied from 38.2 % to 100 %.
The spleens were stored in sterile 1.5 ml Eppendorff tubes and kept at -20 °C until
DNA was extracted from infected spleens using the QIAamp DNA Kit (Qiagen, Clifton Hill,
Australia) according to the manufacturer’s instructions.
The three commercially available Rispens vaccines in Australia (Vaxsafe® RIS,
Poulvac® CVI988 and Nobilis® Rismavac) were obtained from the manufacturers
(Bioproperties, Fort Dodge, Intervet). Prior to phenol-chloroform extraction of DNA, samples
were treated with Proteinase K. All vaccinal samples were subject to sequence analysis of a
fraction of the meq gene to confirm the presence of the targeted polymorphism in meq as
described by Renz (2008). Sequencing of purified DNA was conducted by Macquarie
University, Sydney, Australia using an ABI 377 sequencer (Applied Biosystems Inc., Foster
City, CA, USA). All DNA samples were quantified using a NanoDrop® ND-1000 UV-Vis
spectrophotometer (NanoDrop® Technologies Wilmington, USA). In addition to the six
initial pathogenic MDV1 isolates, the target region of MDV from 14 field samples positive
for MDV1 were sequenced and tested in the two assays.
Extracted DNA was used as template for two standard PCR tests, designed to either
amplify only pathogenic MDV1 or Rispens/CVI988. Primers and the probe for the qPCR
assays were designed using Beacon designer 6.00 (PREMIER Biosoft International, Palo
Alto, USA). The standard PCR was performed in a 25 μl reaction mixture containing 1 μmol
of each primer, 1.8 mM MgCl2, 0.2 mM dNTP’s, 10x reaction buffer (Fisher Biotec, Perth,
Australia), 1 unit of Taq DNA polymerase and approx. 1 ng of template DNA. Amplification
was carried out over 35 cycles each consisting of 1.5 min at 94°C, 1 min at 60°C and 2 min at
72°C, except for the initial 2 cycles in which the period at 94°C was extended to 5 min. After
the final cycle, the elongation phase at 72°C was extended to 10 min with a consecutive step
at 4°C for 5 min. The amplified fragments were separated on an agarose gel (1%) and
visualized by staining with ethidium bromide.
After the primer sets were confirmed to only amplify either pathogenic or
Rispens/CVI988 in several standard PCR assays, TaqMan® real-time qPCR assays were set
up using a RotorGene 3000 real-time PCR machine (Corbett Research, Sydney, Australia).
The qPCR cycling parameters used for the assays were described by Islam et al. (2006).
However, the annealing time was extended from 45 sec to 60 sec.
The sensitivity of the assays was determined by running tenfold serial dilutions of
plasmid DNA with known copy numbers for both pathogenic and vaccinal MDV1. The
lowest dilution in the tenfold dilution series which amplified reliably was defined as the
detection limit. The reproducibility of the qPCR assays with the new plasmid-derived
standard curves was measured by calculating the intra-assay coefficient of variation (CV) by
taking the mean CV for duplicate Ct and calculated copy number for plasmid standards in all
runs of the same assay. The inter-assay CV was determined by comparing the mean Ct and
calculated copy number for each standard in three separate but identical assay runs and
determining the CV for each across assays. Each individual assay was performed on separate
days and serial dilutions of plasmid as well as the reference standards were prepared freshly
on each day. After parallelism with the plasmid standard curves had been confirmed, the
108
Aust. Poult. Sci. Symp. 2010...21
reference standards derived either from spleen tissue or Rispens vaccine were quantified in
terms of viral copy number in three independent identical assays.
III. RESULTS
The product size with both primer sets matched the expected 130 bp (Figure 1). None of the
three Rispens vaccines amplified with the pathogenic specific primer set whereas all six
pathogenic MDV1 isolates amplified (Figure 1, upper lanes). Conversely, none of the six
pathogenic MDV1 isolates amplified with the Rispens specific primer set whereas all three
Rispens isolates did (Figure 1, lower lanes). All 14 additional field isolates which were tested
with both assays only amplified in the pathogenic MDV1 specific assay, but not in the
Rispens specific assay which was expected as all field isolates were derived from flocks
which had not been vaccinated with Rispens (data not shown).
Figure 1:
1% Agarose gel. Upper lanes: pathogenic MDV1 specific primers.
From left to right: λ-HindIII standard, six pathogenic MDV1 isolates,
three Rispens isolates, blank, negative control. Lower lanes:
Rispens/CVI988 specific primers. Same alignment as above.
For the real-time quantitative PCR assay none of the three Rispens vaccines amplified
in the pathogenic MDV assay, but all samples containing pathogenic MDV1 or both Rispens
and pathogenic MDV1 amplified. For the Rispens/CVI988 specific assay pathogenic MDV1
samples did not amplify whereas the 3 Rispens vaccines did.
For both assays, the qPCR conditions were optimised (increased annealing time from
45 to 60 sec) in order to provide sensitivity similar to that of our generic, non-differentiating
MDV1 assay (Islam et al., 2006) which is 2.7 viral copy numbers (VCN) per reaction. With
these settings, standard curves of both tests amplified reliably 10 VCN/reaction. Based on
three individual runs for each assay with samples run in duplicate, intra-assay Ct- values had
a CV of 1.58 % (pathogenic MDV assay) and 1.37 % (Rispens assay), while the mean interassay Ct- values had a CV of 22.34 % and 19.87 % respectively.
109
Aust. Poult. Sci. Symp. 2010...21
IV. DISCUSSION
This is the first set of molecular assays which reliably detect, quantify and differentiate
Australian pathogenic strains of MDV and Rispens, a tool which the poultry industry has
long sought. The tests are highly sensitive and have shown 100% specificity when screened
against the 3 Rispens vaccines on the market and 20 Australian MDVs. They now enable
flock testing for vaccine take, and for monitoring of pathogenic MDV infections in
vaccinated layer and breeder chickens, probably via pooled feather or dust samples. It is now
also possible to test MDV1 isolates for Rispens contamination, and to study the replication
and spread of MDV in breeder and layer populations. To determine the stability of the target
area in our tests over time, the three available Rispens vaccines are currently serially backpassaged five times in chickens and will be subject to DNA sequencing after the last passage.
However it should be noted that these assays are specific for Australian strains of
MDV only. Based on the DNA sequences available online from pathogenic strains of MDV1,
American, Asian or European strains do not differ from Rispens at this position in the meq
gene. We are currently evaluating potential sites of other candidate genes such as pp38, ICP4
or vIL-8 in order to develop assays which will differentiate between both Australian and
international pathogenic MDV1 isolates and Rispens vaccine.
V. ACKNOWLEDGEMENTS
This work is funded by the Australian Egg Corporation Ltd and the Rural Industries Research
and Development Corporation under project AECL 08-17.
REFERENCES
Becker Y, Asher Y, Tabor E, Davidson I, Malkinson M, Weisman Y (1992) Journal of
Virological Methods 40, 307-322.
Churchill AE, Biggs PM (1967) Nature 29, 528-530.
Islam A, Cheetham BF, Mahony TJ, Young PL, Walkden-Brown SW (2006) Journal of
Virological Methods 132, 127-134.
Islam A, Walkden-Brown SW (2007) Journal of General Virology 88, 2121-2128.
Renz KG, Islam A, Cheetham BF, Walkden-Brown SW (2006) Journal of Virological
Methods 135, 186-191.
Renz K (2008) PhD thesis, University of New England, Armidale.
Young P, Gravel J (1996) Current Research in Marek's Disease 5, 308-310.
Schat KA, Calnek BW (1978) Journal of the National Cancer Institue 60, 1075-1082.
Walkden-Brown SW, Tannock GA, Cheetham BF, Islam AFMF, Groves PJ (2006)
Systematic pathotyping of Australian Marek’s disease virus isolates. Final Report of
RIRDC Project UNE-83J, Rural Industries Research and Development Corporation.
Canberra, ACT.
Witter RL, Nazerian K, Purchase HG, Burgoyne, GH (1970) American Journal of Veterinary.
Research 31, 525-538.
Zelnik V, Bernardi G, Wozniakowski G, Philipp H, Rebeski DE (2008) Proceedings of the
8th International Marek’s Disease Symposium. James Cook University. Townsville,
Qld.
110
Aust. Poult. Sci. Symp. 2010...21
A QUANTITATIVE PROFILE OF INFECTIOUS BRONCHITIS VIRUS IN FAECES OF
LAYING HENS
K.K. CHOUSALKAR 1 and J.R. ROBERTS 2
Summary
Two independent real time PCR assays were designed to detect and quantify T and N1/88
strains of infectious bronchitis virus (IBV) from faeces of experimentally-infected
unvaccinated and unvaccinated laying hens. Vaccination VicS-A3-VicS during rearing can
reduce the viral load in the faeces of vaccinated and T or N1/88 challenged hens. Both
vaccinated and unvaccinated hens can shed T strain of IBV in faeces up to 9 weeks p.i. It is
now possible to detect and differentiate N1/88 strain of IBV from vaccine strains. It is also
possible to detect and quantify T and N1/88 strain of infectious bronchitis virus directly from
faeces.
I. INTRODUCTION
Infectious bronchitis virus (IBV) is a highly infectious and contagious pathogen of chickens
worldwide. IBV spreads rapidly amongst the chickens in a flock. The frequency of isolation
varies with the time frame but Alexander and Gough (1977) isolated virus from caecal tonsils
and faeces up to 14wk and 20wks, respectively. Jones and Ambali (1987) isolated the G
strain of IBV up to 28 wks of age in tracheal and cloacal swabs from birds infected at dayold. Currently, reverse transcriptase polymerase chain reaction (RT-PCR) is commonly used
for diagnosis of IBV. Using RT-PCR, IBV has been detected directly from clinical samples
such as the trachea, kidney or cloacal swabs (Mardani et al., 2006). Previously, we have
reported use of the real time PCR assay for the detection and quantification of Australian
strains of IBV from the oviduct (Chousalkar et al., 2009). There are no reports of IBV
detection and quantitation from faeces using molecular diagnostic tests such as real time
PCR. In the current study, the real time PCR test was designed for the rapid detection of IBV
strains from the faeces of IBV infected unvaccinated and vaccinated laying hens.
II. MATERIALS AND METHODS
Day old chickens (n= 190) were obtained from the Baiada Hatchery at Marsden Park. At dayold, all the chickens received Rispens vaccine against Marek’s disease but no other
vaccinations at the hatchery. The chickens were raised on the floor in isolation sheds at the
University of New England. Half the birds were vaccinated with Vic S on day 1 by the
intraocular route at the dose rate of 104.5 embryo infective dose (E.I.D.50). At 4 weeks of age,
birds were exposed to A3 at the dose rate of 103.9 E.I.D.50. At 13 weeks of age, birds were
again vaccinated with Vic S at the dose rate of 104.5 E.I.D.50. Vaccines were obtained from
Fort Dodge, Australia. The other half of the birds remained unvaccinated. At 25 weeks of
age, the unvaccinated and vaccinated birds were divided into two control groups,
unvaccinated unchallenged (UC), vaccinated unchallenged (VC) and four treatment groups,
unvaccinated challenged with T (T), vaccinated challenged with T (VT), unvaccinated
challenged with N1/88 (N) and vaccinated challenged with N1/88 (VN). All the birds were
moved into cages at the age of 25 weeks. At 30 weeks of age, each bird from groups T, VT,
1
2
Charles Sturt University, NSW, Australia
The University of New England, NSW, Australia
111
Aust. Poult. Sci. Symp. 2010...21
N and VN were challenged with one of two different strains of IBV, T and N1/88 (obtained
from Dr. Jagoda Ignatovic, CSIRO, Geelong) and the control birds were sham inoculated
with normal saline. Faeces were collected in individual sterile plastic containers from five
hens from each group at weekly intervals from 1 to 9 weeks post infection (p.i.). RNA from
faeces was extracted as described previously by Culver et al. (2008) using QIAamp DNA
stool mini kit (Qiagen) with some modifications. 0.2 gm faecal samples, collected from 1 to 9
weeks p.i. from five hens of each group, were weighed and dispensed into microcentrifuge
tubes containing 2 mL ASL buffer. The samples were vortexed and heated in a 70°C water
bath for five minutes. The samples were centrifuged at 4800 x g for 10 min and 120 µL of the
superannuant was transferred to another clean microcentrifuge tube containing an inhibitex
tablet. The samples were vortexed and stored at room temperature for 1 min. The samples
were then centrifuged at 4800 x g for 10 min and 200µL of resulting supernatant was treated
with 15µl of proteniase K and 200 µL of AL buffer. The mixture was reheated at70°C and
transferred to a spin column. Washing and elution was performed according to the
manufacturer’s instructions. The elution volume was 100 µL. Extracted RNA was quantified
using Nanodrop and stored at -70°C until used for real time RT-PCR. 50 ng of faecal RNA
was used during the real time PCR reaction. Nucleocapsid sequences from N1/88, A3, Vic S
and T strains of IBV were retrieved from GenBank accession numbers U52599, DQ490205,
U52594 and U52596. All the sequences were aligned using Clustal W2 (EBI sequence
analysis tool, UK). The sequence alignment indicated that the nucleocapsid gene sequences
of A3 and Vic S were dissimilar to N1/88. The primers (Forward primer
5'AGATGGCTGAGCGTAAGTAC-3', Reverse primer 5'-CCTCCTCAATCATCTTGTCATC3') and LNA probe (5’-FAM aaaGgcCaaGctCcaaattt BHQ-2-3’) required for the real time
RT-PCR were generated to amplify the 123 bp sequence within the nucleocapsid region of
the N1/88 strain of IBV. The primers were manufactured by Geneworks (Adelaide, Australia)
and the LNA probe was synthesised by Exiqon (Denmark). The primers and LNA probe for
the real time PCR for the detection of T strain of virus were used as described earlier
(Chousalkar et al., 2009). The LNA probe-based real time reverse transcriptase polymerase
chain reaction (RT-PCR) was developed and performed using a Rotor Gene 3000 real time
PCR machine (Corbett Research, Sydney, Australia) and a one-step RT-PCR kit (Invitrogen
Australia Pty Limited). Raw data were analysed using the default settings of the software for
determination of baseline and threshold of the reaction. The test sensitivity was determined
by running 10 fold serial dilutions of the plasmid DNA (recombinant plasmid DNA /clone)
with known copy numbers. In each assay, a standard curve was generated and used to derive
the infectious bronchitis viral copy number from the oviduct samples. Each sample was
analysed in duplicate.
III. RESULTS
In the N1/88 infected group (N), virus was detected from the faeces of three hens in the first
two weeks, and two hens during the third and fourth weeks. In the VN group, virus was
detected in the faeces of two hens during first week and then in one hen during second and
third weeks. Virus was not detected in either the unvaccinated or vaccinated group from five
weeks p.i.
Amongst the T strain infected group, virus was detected from the faecal samples of
three out of five in the first week and then consistently two out of five hens from two to nine
weeks p.i. Amongst the VT infected group, virus was consistently detected from the faecal
samples of two out of five hens from one to nine weeks p.i. The virus load was low in
vaccinated hens compared to the unvaccinated and challenged hens. The details regarding
the IBV viral load in the faeces are presented in Table 1.
112
Aust. Poult. Sci. Symp. 2010...21
The faecal samples of the hens from control groups were negative throughout the
experiment.
Table 1.
N1/88 and T Viral RNA load from faecal samples of vaccinated and
unvaccinated hen
Weeks
Post
inoculation
1
2
3
4
5
6
7
8
9
Hen
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
N
VN
T
VT
Mean Viral
RNA
11274
9676
123
0
0
9633
10112
193
0
0
676
715
0
0
0
133
192
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Mean Viral
RNA
0
0
1115
136
0
0
0
6639
0
0
0
0
115
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Mean Viral
RNA
886634
14477
758
0
0
216694
1144
0
0
0
12800
1769
0
0
0
8963
16698
0
0
0
1852
7703
0
0
0
2870
3338
0
0
0
928
726
0
0
0
910
109
0
0
0
147
166
0
0
0
Mean Viral
RNA
269089
8790
0
0
0
16694
5279
0
0
0
13462
133
0
0
0
1973
1822
0
0
0
338
432
0
0
0
237
875
0
0
0
494
655
0
0
0
16
20
0
0
0
57
31
0
0
0
113
Aust. Poult. Sci. Symp. 2010...21
IV. DISCUSSION
The real time PCR assay was based on the nucleocapsid region of the N1/88 virus genome
(Sapats et al., 1996). Primers and probes designed within the nucleocapsid region of N1/88
strain of IBV did not give a positive reaction to T, VicS and A3 strains of IBV. Hence it is
now possible to differentiate the N1/88 strain of IBV from vaccine strains. Real-time RTPCR offers both speed and accurate quantification of viral load, and a real time PCR test has
been developed recently using a Taqman probe for the detection of American strains of IBV
from tracheal swabs (Callison et al., 2007) and using LNA probe for the detection of other
Australian IBV strains (Chousalkar et al., 2009). The sequence analysis suggests that the
assay developed earlier by Chousalkar et al. (2009) and Callison et al. (2007) would not
detect the N1/88 strain of IBV used in this study. The current test will be useful to
differentiate and monitor the spread of N1/88 strain of IBV amongst vaccinated chicken
populations. Detection and quantitation of N1/88 and T strains of IBV from the faeces using
real time PCR assay has not been reported previously. The T strain of IBV was isolated from
the faecal samples of T strain infected hens up to 225 days p.i. (Alexander and Gough, 1977).
Massachusetts serotype vaccine virus has been isolated also from cloacal swabs at 63 days
p.i. (Naqui et al., 2003). Excretion of virus and its subsequent isolation from faeces could be
due to the urates present in the faeces. Further studies are required to study the virus detection
simultaneously from kidney, gut contents and faeces.
Virus was consistently isolated from the faecal samples of two hens in the T infected
group from 1 to 9 weeks p.i. Such hens can be persistent virus shedders and these hens
appeared healthy throughout the experiment. Shedding of virus in such fashion by apparently
healthy birds could be an important means of introducing virus into new premises (Jones and
Ambali, 1987). Based on the real time PCR results shown above in the report it could be
concluded that vaccination can reduce the shedding of wild strains (T and N1/88 strains of
IBV) through faeces.
V. ACKNOWLEDGEMENTS
This study was supported by Australian Egg Corporation Limited (AECL). We thank Mr.
Robert Turner for his assistance during the animal trial.
REFRENCES
Alexander DJ, Gough RE (1977) Research in Veterinary Science, 23, 344-347.
Callison SA, Hilt DA, Boynton TO, Sample BF, Robison R, Swayne DE, Jackwood MW
(2007) Journal of Virological Methods, 138, 60-65.
Chousalkar KK, Cheetham BF, Roberts JR (2008) Journal of Virological Methods, 155, 6778.
Culver FA, Britton P, Cavanagh D (2008). Methods in Molecular Biology, 454, 35-42.
Jones RC, Ambali AG (1987) The Veterinary Record, 120, 617-618.
Mardani K, Noormohammadi AH, Ignjatovic J, Browning GF (2006) Avian Pathology, 35,
63-69.
Naqui, S, Gay K, Patalla P, Mondal S, Liu R (2003) Avian Diseases 47, 594-601.
Sapats S, Ashton DF, Wright PJ, Ignjatovic, J. (1996) Virology, 226, 412-417.
114
Aust. Poult. Sci. Symp. 2010...21
THE STRATEGIC USE OF ORGANIC ACIDS TO IMPROVE GUT HEALTH IN
POULTRY
L. LI 1
Summary
Gut health presents significant challenges to modern poultry production. The outcome of
poor gut health, often is costly and devastating to poultry producers. Up to date, numerous
effective solutions have been adopted by producers, in order to improve the gut health. These
actions include: (i) minimising pathogen exposure via bio-security efforts; (ii) improving host
resistance through genetic selection; and (iii) strategic use of feed additives, such as organic
acids, to manipulate intestinal biochemistry to either directly kill or inhibit pathogenic
bacteria colonization and to support the growth of protective bacteria. For decades, organic
acids and salts have been used, as a direct replacement for antibiotic growth promoters
(AGPs) or in combination with AGPs to improve poultry productivity. However, only some
organic acid based products are most likely to make a significant, positive impact on the
poultry industry, as they are developed not only to address current needs, but also to solve the
predictable problems of tomorrow. This paper will focus on the strategic use of organic acids
to improve gut health in poultry.
I. GUT HEALTH SIGNIFICANTLY CHALLENGES MODERN POULTRY
PRODUCTION
Not only does the gut perform such basic functions as digestion of food and absorption of
nutrients, but the gut also functions as a physical barrier to prevent opportunistic pathogens
from colonising and invading host tissues. The gut microflora is a significant part of the gut
mucosal barrier. It is, therefore, not surprising that their activities have a major impact on the
gut health and vice versa. This mutual influence also affects the host, beyond the gut.
Pathogenic bacteria, such as Salmonella, E. coli, Campylobacter and Clostridium, are
commonly found in the gastro-intestinal (GI) tract of healthy birds. Under normal conditions
(good nutrition and adequate immune defence), if a small number of pathogens are present
within the GI tract, the beneficial bacteria can overrule this small number of bad bacteria;
therefore, they may not cause severe clinical infections. The real challenge is when birds
experience stresses, such as changing diet, intestinal disease (virus attach and parasites
infestation) and feed withdrawal, the conditions in the gut might change and then the
opportunistic pathogens will thrive and produce toxins, which in turn, may cause serious
clinic enteric disorders. Generally speaking, birds with clinic enteric disorders have poor
efficiency of feed utilisation, wet droppings, poor uniformity and high mortality. The cost of
any additional treatments will certainly result in more economic loss for the producer.
To prevent economic loss from enteric diseases, AGPs, such as virginiamycin, tylosin
and penicillin etc, have been used extensively in those countries where antibiotics are allowed
to be included in poultry diets. In recent years, with the restrictions on the use of regular
antibiotics, poultry producers have adopted a few effective intervention strategies to reduce
the productivity losses. These strategies include vaccination, possible dietary alternatives to
antibiotics, i.e. organic acids and salts, probiotics and prebiotics. Due to limited space, this
paper will only focus on organic acids and salts, mode of actions and their application
concepts for poultry production.
1
Kemira Asia Pacific, Singapore
115
Aust. Poult. Sci. Symp. 2010...21
II. POSSIBILITY TO USE ORGANIC ACIDS TO IMPROVE GUT HEALTH
From a clinical view point, it is fair to say, that all diseases start in the gut. A healthy gut is
also the key to maximise the nutrient utilisation from feed (Yegani and Korver, 2008). For
decades, poultry feed producers have been aware of organic acid feed supplements as a
means of improving bird health and food safety, but only in a limited capacity. Reluctance for
their widespread use in poultry is mainly due to a limited amount of consistent trial data,
failure to understand their mode of action, and other misguided fallacies.
First of all, short chain fatty acids (SCFAs, C≤ 4), especially propionic and butyric
acid, are essential for the normal structure and function of the gut epithelium. SCFAs are
readily absorbed by the normal intestinal epithelium, and they maintain the integrity of the
intestinal mucosa, stimulate proliferation of normal colonocyte, and stimulate water and
electrolyte absorption from the colonic lumen (Cummings et al., 1987; Velazquez et al.,
1996). The antibacterial effect of dietary SCFAs in poultry takes place mainly in the upper
part of the digestive tract. The reason being that SCFAs are rapidly metabolised and absorbed
from the fore-gut (crop, proventriculus, gizzard) of birds, which will then reduce their
antibacterial activity in the small and large intestines, impacting on growth performance
(Waldroup et al., 1995). Recently, the use of a combination of SCFAs and medium chain
triglycerides (MCTs) as a potential antimicrobial agent has been investigated. This
combination was postulated to be an efficient antimicrobial agent in acidic (gastric-)
environment, as well as in a neutral (intestinal-) environment (Velazquez et al., 1996;
Decuypere and Dierick, 2003; Skrivanova et al., 2005; Van Immerseel et al., 2004). The
mode of action of the synergetic antimicrobial effect can be summarised as: Most pH
sensitive bacteria do not grow in acidic environment as acid tolerance is a major factor that
limits how well certain types of bacteria survive in an ecosystem. SCFAs reduce fore-gut pH,
hence increase barrier function for harmful microbes and improve nutrients digestion.
Simultaneously, MCFAs from medium chain triglycerides are liberated in the intestines by
intestinal lipases, providing a natural “slow release” antimicrobial function. The final
outcome is improved gut health (low wet dropping score) and enhanced bird performance.
Furthermore, the reduced microbial growth in the intestines, allows the birds to more
efficiently utilise available nutrients.
Numerous studies have shown that SCFAs and medium chain fatty acids (MCFAs)
and medium chain triglycerides (MCTs) are effective in either directly killing or inhibiting
Salmonella, Campylobacter and Clostridium perfringens colonization and supporting the
growth of protective Bifidobacteria and Lactobacilli bacteria in poultry (Dibner and Buttin
2002; Owens et al., 2008; Solis de los Santos et al., 2009). More importantly, studies
demonstrated that supplementation of feed or drinking water with organic acids could be used
as potential additives for the control of necrotic enteritis (Gornowicz and Dziadek, 2002;
McDevitt et al., 2006). Therefore, organic acids not only control pathogenic bacteria in the
feed and GI tract of the host, they also help the bird to achieve maximum nutrient digestion
and absorption, hence, improved productivity.
Formulating a diet with a low acid binding capacity (ABC) diet to reduce diarrhoea
incidence is a practical consideration. Most pH sensitive pathogens, such as Salmonella, E.
coli and Campylobacter, stop growing in the environment with a pH lower than 4.5. It is
generally established that the gastric acid secretion is proportional to the buffering capacity of
the meal. On the other hand, the secreted acids must be neutralised in the duodenum by the
bicarbonates from the bile. This will dilute the digestive enzymes secreted by the pancreas
and increase the feed flow, leading to malabsorption and diarrhoea (Decuypere et al., 1997).
It is, however, neither practical, nor economical to achieve this low feed pH by adding large
quantities of acids. In practice, feeds with low ABC value (meq/kg) can be formulated by: (i)
116
Aust. Poult. Sci. Symp. 2010...21
lowering the calcium levels to the minimum required, (ii) replacing inorganic calcium with
organic calcium sources, (iii) restricting the inclusion of inorganic phosphates using phytase
enzymes and (iv) formulating feeds with lower crude protein levels.
III. CONCLUSION
Successful use of organic acids in poultry production requires knowledge of their mode of
action and their numerous potential benefits. One of the most important benefits is to lower
feed and gastric pH, thereby, reducing the buffering capacity of feeds and enhancing nutrient
utilisation. Furthermore, organic acids reduce intestinal colonisation of pathogens; affect the
composition of intestinal microflora and mucosal morphology. In practice, a combination of
short chain fatty acids and medium chain fatty acids can be applied along with nutritional,
management and biosecurity measures in order to improve gut health in poultry.
REFERENCES
Cummings JH, Pomare EW, Branch WJ, Naylor CPE, Macfarlane GT (1987) Gut 28, 12211227.
Decuypere JA, Dierick NA (2003) Nutrition Research Reviews 16, 193-209.
Dibner JJ, Buttin P (2002) The Journal of Applied Poultry Research 11, 453-463.
Gornowicz E, Dziadek K (2002) Annals of Animal Science 2 (Suppl. 1), 93-96.
McDevitt RM, Brooker JD, Acamovic T, Sparks NHC (2006) World's Poultry Science
Journal 62, 221-247.
Owens B, Tucker L, Collins MA, McCracken KJ (2008) British Poultry Science 49, 202-212.
Skrivanova E, Marounek M, Dlouha G, Kanka J (2005) Letters in Applied Microbiology 41,
77-81.
Solis de los Santos F, Donoghue AM, Venkitanarayanan K, Metcalf JH, Reyes-Herrera I,
Dirain ML, Aguiar VF, Blore PJ, Donoghue DJ (2009) Poultry Science 88, 61-64.
Van Immerseel F, De Buck J, et al. (2004) Applied and Environmental Microbiology 70,
3582-3587.
Velazquez OC, Seto RW, Rombeau JL (1996) Proceedings of the Nutrition Society 55, 49-78.
Waldroup A, Kaniawati S, Mauromoustakos A (1995) Journal of Food Protection 58, 482489.
Yegani M, Korver DR (2008) Poultry Science 87, 2052-2063.
117
Aust. Poult. Sci. Symp. 2010...21
INACTIVATION OF VIRUSES AND COCCIDIA IN BROILER LITTER FOLLOWING
HEAPING OR WINDROWING AT THE END OF THE BATCH
A.F.M.F. ISLAM 1, S.K. BURGESS1, P. EASEY 2, B. WELLS 3 and
S.W. WALKDEN-BROWN1
Summary
Two on-farm experiments were conducted to determine the effectiveness of windrowing or
heaping end of batch broiler litter for up to 10 days on inactivation of pathogenic viruses and
coccidia. In Expt 1 (Sydney) litter treatments were heaping, heaping with turning at day 4 or
windrowing. In Experiment 2 (Brisbane) litter treatments were no heaping (simply turning
litter in situ), windrowing, or windrowing with turning at day 4. There were two replicates of
both treatments in each experiment. On 4 occasions (days 0, 3, 6, 9 in Expt. 1 and days 0, 4, 7
and 10 in Expt. 2) representative litter samples from each treatment were subjected to a chick
bioassay to measure litter infectivity for a number of key viral diseases as determined by
seroconversion at 35 days post exposure to the litter. Coccidial oocyst counts in faeces were
also conducted in Expt 1. Heaping or windrowing litter led to marked reduction in the
proportion of bioassay chicks positive for chicken anaemia virus (CAV) and fowl adenovirus
(FAV) but there was no clear effect on infectious bursal disease virus (IBDV) for which
initial litter infectivity was low. Turning provided no additional benefit overall and heaping
appeared to provide a greater level of inactivation than windrowing. Litter infectivity was
very low or absent for pathogens such as infectious bronchitis virus (IBV) and Marek’s
Disease virus (MDV) so inferences could not be made about these. Serology for additional
viral pathogens is ongoing. Coccidial oocysts were completely inactivated in the treated litter
by day 6. These preliminary data suggest that heaping or windrowing of litter is beneficial in
reducing viral pathogen load in litter, that most (but not all) inactivation is completed by days
6-7, that large heaps inactivate more effectively than windrows, and that turning of heaps or
windrows does not provide an additional benefit.
I. INTRODUCTION
Re-use of litter by broiler chickens can reduce the environmental impact and cost of chicken
production but uptake of the practice is limited by risks of pathogen carryover. Unlike some
other major poultry producing countries, only a small proportion of Australian broiler farms
reuse litter in the shed (East, 2007). The reluctance to reuse litter in the Australian broiler
industry is primarily based upon concerns on the animal health and productivity issues
(Groves, 2003) particularly due to carry-over infection with pathogens, particularly viruses.
A small proportion of Australian broiler growers do reuse litter in the shed following partial
composting by heaping or windrowing for 4-10 days, where a temperature of ~60˚C can be
reached inside the litter heap or windrow. The aim of this study was to assess the
effectiveness of such litter treatments over time on the inactivation of selected poultry viruses
and coccidia under field situations. Unlike bacteria, viruses are not easy to detect and
quantify from poultry litter, as viruses do not grow in non-living media. Therefore, little or no
data are available on viral survival rate in poultry litter following different litter treatments.
Recently we developed and validated a bioassay to detect and quantify viral pathogens from
1
School of Rural and Environmental Science, University of New England, Armidale NSW 2351
Cordina Chicken Farms, Gretystanes NSW 2145
3
Wells Avian Consultancy, Glenorie BNSW 2157
2
118
Aust. Poult. Sci. Symp. 2010...21
poultry litter (Islam et al., 2009) and this tool was used to measure change in litter infectivity
following treatment in the present study.
II. MATERIALS AND METHODS
Two field experiments were conducted. Experiment 1 had three litter treatments in duplicate
in the three sheds of a Sydney region farm. Each shed was divided into half to make six
experimental units. Three litter treatments were applied in the six units in duplicate for nine
days. The treatments involved forming half of the litter in the shed into windrows (W), a
long row of litter approx. 1.0–1.2 m tall, 2 m wide and 40 m long along the centre of the
shed, heaps (approx 2.5 m in height) without turning (H) of with mechanical turning on day 3
(HT). Representative litter were collected and transported to UNE at days 0, 3, 6 and 9 for
bioassay of litter infectivity as described by Islam et al. (2009). Experiment 2 was conducted
in a Brisbane region farm, where three treatments; windrow (W), windrow with turn (WT)
and no heaping (NH) with litter left alone and mechanically turned at day 4, were applied in
six different sheds in duplicate. Litter was collected at days 0, 4, 7 and 10 for bioassay.
The experimental bioassay utilized a 3×4 factorial design with three treatments and four
time periods as described above for each field experiment. Each cell was replicated in 2
isolators, and there were also 2 negative control isolators. In each isolator 10 specific
pathogen free chickens were exposed to 8 L of litter. At day 35 post-exposure serum samples
were collected and chickens held until a common age across treatments was reached (42 days
old) after which they were humanely killed, weighed and sampled for spleen and faeces.
Sera were tested for antibodies directed against IBV (ELISA), CAV (ELISA), IBDV
(ELISA), Infectious laryngotracheitis virus (ILTV, ELISA) and FAV (ELISA). MDV was
detected using real-time quantitative PCR of DNA extracted from spleen. MDV was
quantified in spleen samples using real-time quantitative PCR and oocysts counted using
standard methods. Data were analysed using appropriate models in JMP Version 7.
III. RESULTS
Negative control chickens were negative for all the pathogens tested in both experiments.
Serological data for experiments 1 and 2 are summarised in Tables 1 and 3 respectively. The
field litter collected prior to the application of treatments (day 0) was highly positive for
CAV and FAV in both experiments. Positive samples for IBDV (both experiments) and IBV
(experiment 2) were also observed in some treatment groups, but at much lower levels
(Tables 1 and 2). None of the chickens was positive for ILTV or MDV in experiment 1 (data
currently not available for experiment 2).
Table 1.
Treatment
Neg. control
Pre treat (d0)
Heap No Turn
(H)
Heap Turn
(HT)
Windrow (W)
Experiment 1. Proportion and % (in brackets) of SPF chickens serologically
positive for CAV, FAV IBDV and IBV 35 days after exposure to treated litter.
Day posttreatment
0
3
6
9
3
6
9
3
6
9
CAV
FAV
IBDV
0/17 (0)
29/30 (97)
7/19 (36)
1/19 (5)
0/19 (0)
19/19 (100)
8/20 (40)
0/20 (0)
14/20 (70)
10/18 (55)
4/20 (20)
0/17 (0)
20/27 (74)
0/19 (0)
0/20 (0)
0/18 (0)
0/19 (0)
0/19 (0)
8/21 (38)
0/20 (0)
1/19 (5)
0/21 (0)
0/17 (0)
7/33 (21)
2/19 (11)
11/20 (55)
0/18 (0)
1/21 (5)
5/20 (25)
2/20 (10)
1/20 (5)
1/20 (5)
0/21 (0)
119
IBV
0/17
0/30
0/22
0/20
0/18
0/19
0/19
0/20
0/18
0/20
0/21
(0)
(0)
(0)
(0)
(0)
(0)
(0)
(0)
(0)
(0)
(0)
Aust. Poult. Sci. Symp. 2010...21
Table 2.
Experiment 2. Proportion and % (in brackets) of SPF chickens serologically
positive for CAV, FAV IBDV and IBV 42 days after exposure to treated litter.
Treatment
Day posttreatment
Neg. control
Windrow
No
Turn (W)
Windrow
(WT)
0
4
7
10
0
4
7
10
0
4
7
10
turn
No heaping
CAV
FAV
IBDV
IBV
0/17 (0)
13/20 (65)
1/18 (6)
0/20 (0)
0/20 (0)
19/20 (95)
3/18 (17)
1/20 (5)
1/20 (5)
19/20 (95)
15/18 (83)
8/20 (40)
7/20 (35)
0/17 (0)
15/20 (75)
8/18 (44)
0/20 (0)
11/20 (55)
19/20 (95)
9/18 (50)
0/20 (0)
0/20 (0)
19/20 (95)
17/18 (94)
10/20 (50)
9/20 (45)
0/17 (0)
1/20 (5)
1/19 (5)
0/20 (0)
0/20 (0)
5/20 (25)
1/19 (5)
0/20 (0)
0/20 (0)
0/20 (0)
0/18 (0)
2/20 (10)
0/20 (0)
0/20 (0)
0/20 (0)
0/20 (0)
0/20 (0)
0/20 (0)
0/20 (0)
0/20 (0)
2/20 (10)
0/20 (0)
0/20 (0)
0/20 (0)
2/20 (10)
0/20 (0)
Heaping or windrowing litter led to marked reduction in the proportion of bioassay chicks
positive for chicken anaemia virus (CAV) but infectivity remained after 9-10 days in the W
(but not H) treatments (Tables 1 and 2). For FAV there was also a marked reduction,
particularly in Expt 1, but there was unexpected reappearance of the pathogen in samples
following a clear test and in both experiments there were some positive samples at days 9-10.
The results for IBDV are inconclusive as initial litter infectivity was low but again positive
samples were observed as late as day 9. Overall, turning provided no additional benefit and
heaping appeared to provide a greater level of inactivation than windrowing. In Expt. 1
coccidial oocyst counts were dramatically reduced by litter treatment and no chicken exposed
to day 6 and 9 litters was positive for coccidia (Table 3).
Table 3.
Experiment 1. Coccidial oocyst count (per gram of faeces) from pooled faecal
samples collected at 42 days post exposure to litter.
Shed
Treatment
Oocyst count (g/faeces)
Day 0
Heap Turn
Windrow
Heap NT
B
Heap Turn
Heap NT
4
Windrow
Neg Control Nil
A
241,000
42,500
16,000
Day 3 Day 6 Day 9
0
0
0
1,000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
IV. DISCUSSION
The results demonstrate clear reductions in the infectivity of litter for CAV, FAV and
coccidia with heaping and windrowing treatments. Coccidial oocysts were inactivated
quickly, with no evidence of litter transmission at day 6 onward. This is consistent with the
report that high temperature (~65˚C) in the presence of ammonia inactivates coccidial oocysts
within hours (McDougald and Fitz-Coy, 2008).
120
Aust. Poult. Sci. Symp. 2010...21
CAV and FAV are relatively environmentally resistant viruses (Schat and Woods, 2008;
Adair and Fitzgerald, 2008) and the reduction in infectivity was greatest for the heap
treatment without turning in Expt 1. This treatment resulted in no detectable infectivity by
day 9. Turning the heaps appeared to reduce the rate of decline in infectivity and resulted in
some infectivity appearing at day 9 for FAV. Windrow treatments clearly reduced CAV and
FAV infectivity, in some cases removing infectivity as early as day 7 (Expt 2). Turning the
windrow in Expt 2 reduced its efficacy against CAV, but not FAV for which the turned
windrow was superior. Overall detectable infectivity remained at days 9-10 in 2 of 3 windrow
treatments for CAV and 1 of 3 windrow treatments for FAV.
The results for IBDV are inconclusive as initial litter infectivity was low but again
positive samples were observed as late as day 9 demonstrating the well-documented
environmental resistance of this virus. The appearance of IBV in two samples at day 7 posttreatment is unexpected as the virus is fragile and we have previously found that it does not
transmit effectively on litter particularly after transportation (Islam et al., 2009).
Two limitations to this study are apparent. The first is that the outcome is dependant on
the range of pathogens present on the site being tested. Thus we have limited or no data on
the effectiveness of the litter treatments against MDV, ILTV and IBDV due to the low level
of initial pathogen load in the samples. In future it may be preferable to work with litter that
has been deliberately contaminated by chickens harbouring a wider range of pathogens. A
second limitation is the appearance of unexpected increases in infectivity of samples over
time. This occurred in some cases with FAV, IBDV and IBV but and were mostly small. It
probably reflects sampling variation despite the rigorous sampling procedure used (10 sites
and 4 depths per windrow or heap). A localised source of surviving virus in the litter may
only be sampled by chance on one of the sampling days. A small number of weak positives
following a series of negative results may also represent false positives. Expressing the data
in terms of mean titre, rather than ratio of positive to negative may reduce this effect.
Despite this, the study has provided a clear demonstration of the efficacy of litter heaping
and windrow treatments in reducing litter transmission of CAV, FAV and coccidia.
V. ACKNOWLEDGEMENTS
This work was funded by the Australian Poultry CRC Project 06-18. We are grateful to the
poultry producers and company technical services staff involved in this study.
REFERENCES
Adair BM, Fitzgerald SD (2008) In: Diseases of Poultry 12th edition. pp 252-266. Iowa State
University Press. USA
East IJ (2007) Australian Veterinary Journal 85, 107–112
Groves PJ (2003). Poultry News, Rural Press, Australia, pp 1 and 31.
Islam AFMF, Walkden-Brown SW, Groves PJ, Wells B (2009) Proceedings, Australian
Poultry Science Symposium 20, 176-179.
McDougald LR, Fitz-Coy SH (2008) In: Diseases of Poultry 12th edition. pp 1068-1085. Iowa
State University Press. USA. 1068-1085.
Schat KA, Woods LW (2008) In: Diseases of Poultry 12th edition. pp 209-249. Iowa State
University Press. USA.
121
Aust. Poult. Sci. Symp. 2010...21
SPATIAL AND TEMPORAL VARIATION IN AMMONIA CONCENTRATIONS IN
BROILER SHEDS: EFFECTS OF CHICKEN AGE AND SHED TYPE
A.F.M.F. ISLAM 1, M. DUNLOP 2, B. WELLS 3, and S.W. WALKDEN-BROWN1
Summary
For studies monitoring ammonia concentrations in chicken sheds, determining a
suitable position and height of the monitoring equipment in a shed is important. To
investigate this an experiment was conducted in which aerial ammonia concentrations were
measured at 10 positions within the shed and three heights (5, 30 and 150 cm above the litter)
in conventional and tunnel ventilated sheds at various chicken ages throughout the batch on 8
farms in the Mangrove Mountain and Sydney regions during autumn. All sheds contained
chickens on new litter following full cleanout. Position of measurement within the shed had
no significant effect on ammonia concentrations, but height above the litter did, with
significantly higher ammonia at 5 cm than other heights. Ammonia concentrations remained
within acceptable limits (< 25 ppm) throughout, increasing rapidly with chicken age up to
week 3 then holding and reducing after week 5. Slightly higher ammonia concentrations were
observed in conventional than tunnel ventilated sheds.
I. INTRODUCTION
Ammonia (NH3) is a colourless irritant alkaline gas produced during the decomposition of
organic matter by bacterial de-amination of nitrogenous substances in the poultry shed. High
NH3 is detrimental to chicken welfare and productivity, causing reduced feed intake and
growth and increased susceptibility to pathogens (Becker et al., 2002; Miles et al., 2004; Ritz
et al., 2004). The reported concentrations at which adverse affects occur vary but an exposure
limit of 25 ppm has generally been set by industry and welfare bodies of different countries
on human safety and animal welfare grounds (Kristensen and Wathes, 2000; Wathes et al.,
1983). Ammonia concentrations in broiler sheds depend on many factors including litter
moisture level, pH, ventilation rate and chicken age. In-shed NH3 could also vary depending
on location within a shed and this effect may differ between conventional and tunnel
ventilated sheds. We plan to conduct several longitudinal studies of ammonia production in
the field and the issue of where to locate the monitoring equipment within the shed is
important. This study was therefore conducted to determine the spatial and temporal variation
of NH3 in both types of broiler sheds at various chicken ages to resolve this issue.
II. MATERIALS AND METHODS
The experiment was conducted on 8 co-operating broiler farms in Mangrove Mountain and
north-western Sydney basin area between 26 March and 4 April 2009 and details are given in
Table 1. Aerial ammonia concentrations in ppm were measured using VRAE7800 Hand Held
Gas Surveyors (Geotech Environmental Equipment, Inc., Colorado, USA) calibrated against
a 50 ppm NH3 standard sample. Wind speed, humidity and temperature were recorded with
1
School of Rural and Environmental Science, University of New England, Armidale NSW 2351
Queensland Primary Industries and Fisheries, Toowoomba QLD 4350
3
Wells Avian Consultancy, Glenorie NSW 2157
2
122
Aust. Poult. Sci. Symp. 2010...21
Kestrel Weather Meter K4000 (Nielsen-Kellerman, Inc., PA, USA). All equipment had a data
logging function.
The experiment was a 2 × 4 factorial design, with two types of broiler shed (conventional and
tunnel ventilated) and four ages of chickens (Weeks 1, 2-3, 5 and 7). All selected sheds were
on different farms and all used fresh wood shavings as litter material, following full cleanout.
Within the overall design there were 3 separate studies viz:
1. Position within shed. Between 1130 and 1400 h, measurements were taken at a fixed
height of 30 cm from 10 fixed positions within the shed in both the morning (between 900
and 1130 h) and afternoon (between 1400 and 1600 h). Ammonia was measured for at
least 15 minutes at 3-minute intervals at each location with temperature and wind speed
also averaged over this period. The order of measurement of each location was selected at
random each time.
2. Height above the litter. Measurements were taken from nine positions at heights of 5, 30
and 150 cm above floor level. Again measurements were for 15 minute periods or more.
3. Overnight fluctuation. Between approximately 1700 and 0700 h continuous
measurements were recorded at two positions and two heights (30 and 150 cm).
Table 1.
Details of the experimental sheds and chickens.
Sample
date
Shed
type
Shed
size (m)
Ventilation
system
Chicken
age (days)
Sex
Number
of birds
27/03
Conv.
70 × 14
Blinds down
7
Mixed
21,000
28/03
Tunnel
160 × 14
1 x low speed fan
6
Male
40,000
30/03
Conv.
120 × 16
Blinds down
14
Mixed
29,400
31/03
Tunnel
100 × 14
Fans on auto
22
Mixed
21,800
01/04
Tunnel
120 × 15
Fans on auto
33
Mixed
29,500
02/04
Conv.
100 × 13
Completely open
34
Mixed
17,600
03/04
Tunnel
138 × 16
Fans on auto
45
Male
4,000 (17,000)
04/04
Conv.
112 × 13
Completely open
46
Mixed
17,000 (22,000)
For studies 1 and 2, data were averaged for each position and height to provide a
single estimate for analysis. Position data (Study 1) were analysed by ANOVA for the fixed
effects of shed type, position, period (morning and afternoon) and chicken age. Height data
(Study 2) were analysed by ANOVA and for the fixed effects of shed type, chicken age,
position in the shed and height above the litter. Overnight data (Study 3) were not analysed
statistically. Data for each time point were plotted against time to show the NH3 level in each
shed.
III. RESULTS
In study 1 (Position) there was a significant effect of age (P < 0.0001) on the ammonia
concentration at 30 cm above the floor without a significant effect of shed type (P = 0.08),
position (P = 0.78) or period (P = 0.09). There were no significant interactions between these
effects. The mean NH3 concentrations in both types of sheds at various ages of chickens are
shown in Table 2.
123
Aust. Poult. Sci. Symp. 2010...21
In study 2 (Height) there were significant effects of height (P < 0.0001), shed type (P
< 0.001) and age (P < 0.0001) but not position on NH3 concentrations. There were also
significant interactions between the effects of age and height (P < 0.0001) but not height by
shed type (P < 0.061) or other interactions. Ammonia concentrations were higher in
conventional than tunnel ventilated sheds (9.5 vs 8.4 ppm). No height effect was evident up
to week 3, after which NH3 were significantly higher at 5 cm than any other heights (Figure
1).
Table 2.
Study 1. Mean ammonia concentration, temperature, relative humidity and wind
speed in tunnel ventilated and conventional broiler sheds containing chickens of
various ages.
Tunnel ventilated shed
Age
(d)
NH3
(ppm)
6
0.81
22
12.16
33
45
Figure 1.
Temp
(˚C)
Conventional shed
RH
(%)
Max wind
sp. (km/h)
Age
(d)
NH3
(ppm)
Temp
(˚C)
RH
(%)
Max wind
sp. (km/h)
42-73
1.3
7
2.17
19-32
49-93
1.8
20-26
74-100
1.8
14
6.31
19-28
56-94
1.4
12.25
19-27
60-93
3.3
34
12.57
18-25
54-83
3.0
9.26
20-25
92-100
6.0
46
9.82
15-24
74-95
4.4
19-32
Study 2. Interaction between the effects of height of measurement above the floor and
chicken age on air ammonia concentrations.
The overnight data (Study 3) were plotted against time for shed types (Fig 2). There was not
much difference in ammonia concentration in conventional and tunnel ventilated sheds,
however, the variability of air ammonia in the conventional sheds appeared to be greater than
in tunnel ventilated sheds. There was also no discernable systematic variation in ammonia
concentrations over the period of measurement.
IV. DISCUSSION
The main purpose of this study was to evaluate spatial variation of NH3 concentrations in
sheds to determine whether single point monitoring in longitudinal studies would be valid.
124
Aust. Poult. Sci. Symp. 2010...21
There was no systematic difference in ammonia concentrations for the chosen 10 positions
within the shed indicating that that monitoring of NH3 concentrations from any of these
positions within the shed is representative of the whole shed. However, there was a
significant effect of height above the litter, with ammonia concentrations being higher near
the surface of the litter (5 cm). The height at which sampling occurs should be selected to
represent the air being inhaled by the target population (chickens or people).
Figure 2.
Study 3. Overnight air ammonia concentrations in conventional and tunnel ventilated
broiler sheds containing chickens of various ages. Mean of two heights (30 and 150
cm).
The ambient NH3 concentrations in the eight studied broiler farms were relatively low
compared to some overseas studies (eg. Wathes et al., 1997), where in-shed NH3
concentrations often exceed 50 ppm. Mean HN3 concentrations always remained below the
threshold value of 25 ppm, with slightly higher NH3 concentrations in conventional than
tunnel ventilated sheds. This may reflect the use of new litter in this study as NH3 could be
expected to be higher on reused litter (Miles et al., 2004).
Air NH3 concentrations increased with increasing age of the chickens to week 3 then
plateaued to week 5 before declining. The plateauing after week 3 is likely due to increased
ventilation rates after brooding. The decline after week 5 is likely due to both increased
ventilation rates and decreased stocking densities following partial removal of chickens.
Overall, we conclude that NH3 concentrations in Australian broiler production systems based
on full cleanout and single use litter are relatively low and the distribution within a shed is
homogenous in regard to position, but not height.
ACKNOWLEDGEMENTS
This work was funded by the Australian Poultry CRC under Project 06-15. We thank Mark
Johnstone of Cordina Chicken Farms for help during the field measurements
REFERENCES
Becker A, Vanhooser SL, Swartzlander JH, Teeter RG (2004) Applied Poultry Science 13, 5-9.
Kristensen HH, Wathes CM (2000) World’s Poultry Science Journal 56, 235-45.
Miles DM, Branton SL, Lott BD (2004) Poultry Science 83, 1650-54.
Ritz CW, Fairchild BD, Lacy MP (2004) Journal of Applied Poultry Research 13, 684-692.
Wathes CM, Holden MR, Sneath RW, White RP, Philips VR (1997) British Poultry Science 38, 1428.
125
Aust. Poult. Sci. Symp. 2010...21
EGGSHELL FACTORS INFLUENCING EGGSHELL PENETRATION AND WHOLE
EGG CONTAMINATION BY DIFFERENT BACTERIA, INCLUDING SALMONELLA
ENTERITIDIS
K. DE REU 1, W. MESSENS1, K. GRIJSPEERDT1, M. HEYNDRICKX1,
M. UYTTENDAELE 2 and L. HERMAN1
Summary
Trans-shell infection routes and whole egg contamination of seven phylogenetically diverse
bacterial strains were studied. Eggshell penetration was correlated with various eggshell
characteristics and the identity of strains. Contrary to the cuticle deposition, the shell surface
area, shell thickness and number of pores did not influence the eggshell penetration. The
results indicate that Gram-negative, motile and non-clustering bacteria penetrated the
eggshell most frequently; Pseudomonas sp. (60%) and Alcaligenes sp. (58%) were primary
invaders followed by Salmonella Enteritidis (43%). In comparison with the non-Salmonella
strains, Salmonella Enteritidis was a primary invader of whole eggs (32%). Egg related
Salmonella Enteritidis strains had no special capacity to contaminate whole eggs. Penetrated
eggshells and contaminated whole eggs showed a significant higher bacterial eggshell
contamination.
I. INTRODUCTION
Transovarian or “vertical” transmission of micro-organisms occurs when eggs are infected
during their formation in the hen’s ovaries. Horizontal transmission occurs when eggs are
subsequently exposed to a contaminated environment and micro-organisms penetrate the
eggshell. Studies conducted by Barrow and Lovell (1991) suggest that most of the
contamination is due to horizontal transmission, although others do not agree (Humphrey,
1994). Contents contamination of whole intact eggs with Salmonella Enteritidis should be
mainly the result of infection of the reproductive tissue (Humphrey, 1994). In this study the
influence of eggshell characteristics on the eggshell penetration on the one hand and the egg
content contamination on the other hand was investigated. To study more in detail the
potential of Salmonella to contaminate whole eggs by the horizontal infection route, whole
egg contamination with different Salmonella strains was determined.
II. MATERIALS AND METHODS
Intact eggs (no cracks, pin-holes) were filled with agar and inoculated (agar approach) or
directly inoculated (intact egg approach). Seven phylogenetically diverse bacterial strains;
Staphylococcus warneri, Acinetobacter baumannii, Alcaligenes sp., Serratia marcescens,
Carnobacterium sp., Pseudomonas sp. And Salmonella Enteritidis, all own isolates from egg
contents, were inoculated in the first study. In the second study four different Salmonella
Enteritidis strains and one Salmonella Typhimurium strain were used. The four Salmonella
Enteritidis strains were originally respectively isolated from two different egg contents, from
a deer and a lizard; the Salmonella Typhimurium strain was isolated from overshoes taken at
1
Institute for Agricultural and Fisheries Research, Technology and Food Science Unit,
Brusselsesteenweg 370, 9090 Melle, Belgium.
2
Department of Food Safety and Food Quality, Laboratory of Food Microbiology and Food
Preservation, University of Ghent, Coupure Links 653, 9000 Ghent, Belgium
126
Aust. Poult. Sci. Symp. 2010...21
the outside environment of a pig farm. Strains were selected for resistance to streptomycin.
An agar method (agar approach) described by Berrang et al. (1998) was adapted to study and
visualize the bacterial eggshell penetration. The used method is described in detail by De Reu
et al. (2006). Where bacterial penetration occurred, organisms grew on the agar and reduced
the indicator triphenyl tetrazolium chloride to formazan which is red in color. Penetration was
recorded when red colonies on the agar were visible by candling. Agar-filled (agar approach)
and whole eggs (intact egg approach) were inoculated by immersion for 1 min in phosphate
buffered saline containing 105–106 CFU/ml of a streptomycin resistant strain of one of the
selected species. This resulted in 103–104 CFU of the selected bacterium on the eggshell.
After drying, the eggs were stored at 20°C and 60% relative humidity for up to 21 days. At
day 0 and day 21 the eggshell contamination with the selected strains was determined. To
remove the egg contents of whole eggs aseptically (intact egg approach), a modification of
the method described by Himathongkham et al. (1999) was used. During the eggshell
penetration and egg content contamination experiment of the first study, different eggshell
characteristics were determined. The shell surface area, the shell thickness, the number of
pores, and the cuticle score were studied in the penetration experiment. In the whole egg
contamination experiment only shell surface area and loss of weight at the pores were
measured. Details on the used methods are outlined by De Reu et al. (2006).
III. RESULTS
No significant difference between area eggshell, shell thickness and number of pores and the
presence or absence of bacterial eggshell penetration was found. The mean cuticle deposition
was lower for penetrated compared to non-penetrated eggshells (individual strain and all
strains). For the individual strain Carnobacterium sp. And for the general result of all strains
this difference was significant (P < 0.001). The whole egg contamination was not influenced
by either the area of the eggshell or by the porosity of the eggshell.
The individual data per selected strain and the general data (all bacterial strains)
obtained with the agar approach, showed a higher count of the inoculated strain on the
eggshell at day 21 for penetrated eggshells compared to non-penetrated eggshells. This higher
count was even significant for the general data (P < 0.001) and for six of the seven selected
strains; respectively for S. warneri, Alcaligenes sp., A. baumannii, Pseudomonas sp.,
Salmonella Enteritidis and S. marcescens (respectively P = 0.011, < 0.001, 0.0018, < 0.001,
0.0016 and 0.0038). The count of bacteria on the entire shell of whole eggs was on average
0.6 log CFU lower compared to agar-filled shells; respectively 1.7 versus 2.3 log CFU/shell.
For 5 of the 7 selected strains the contaminated whole eggs had a (slightly) higher count of
the inoculated strain on the eggshell at day 21; for none of the individual strains this was
significant. The overall data of all strains showed that the count on the eggshell of the
contaminated whole eggs was significantly higher (P = 0.0029); 1.89 log CFU/shell versus
1.66 log CFU/shell for the non-contaminated whole eggs.
Figure 1a shows the percentage of eggshell penetration (agar approach) for all strains
tested, after 21 days of incubation. Pseudomonas sp. and Alcaligenes sp followed by
Salmonella Enteritidis penetrated the eggshell most frequently. They accounted for 60, 58
and 43% of the agar-filled eggs penetration, respectively. Figure 1b shows the percentages of
whole egg contamination (intact egg approach). The egg contents of whole eggs were most
frequently contaminated by Salmonella Enteritidis (33%) followed by Carnobacterium sp.
(17.5%). All strains were able to penetrate in agar-filled eggs (eggshell penetration) as well as
to contaminate whole eggs (whole egg contamination). Of the 403 agar-filled eggs, 131
(33%) were penetrated by the selected strains compared to a content contamination of 16%
(60 on 385) whole eggs.
127
Aust. Poult. Sci. Symp. 2010...21
For the second study; contamination percentages of 18%, 6%, 14% and 26% for
Salmonella Enteritidis isolated respectively from two different egg contents, a deer and a
lizard; and of 24% for Salmonella Typhimurium isolated from overshoe of a pig house were
found. Fifty intact whole eggs were used in each case. Average eggshell contaminations on
day 21 were comparable.
90
Median
25%-75%
Non-Outlier Range
Outliers
80
90
70
60
60
50
50
40
40
30
30
20
20
Whole egg c
Eggshell penet
70
10
10
0
Median
25%-75%
Non-Outlier Range
Outliers
80
S. w arneri
Alcaligenes sp. Pseudomonas sp. S. marcescens
Carnobacterium sp. A. baumannii Salmonella Enteritidis
Figure 1a:
Figure 1b:
0
S. w arneri
Alcaligenes sp. Pseudomonas sp. S. marcescens
Carnobacterium sp. A. baumannii Salmonella Enteritidis
Percentage of eggshell penetration for each individual bacterial strain.
Percentage whole egg contamination for each individual bacterial strain.
IV. DISCUSSION
In our study a significant lower cuticle deposition was found on penetrated eggshells
compared to non-penetrated eggshells. Alls et al. (1964) found that cuticle removal increased
microbial contamination from 20 to 60%. Drysdale (1985) found also a significantly higher
bacterial contamination in eggs which had a poor cuticle (40%) compared to eggs with a
medium or good quality cuticle (26%). The defence of the cuticular layer has on the other
hand been questioned by Nascimento et al. (1992) and Messens et al. (2005) using agar-filled
eggs.
A positive correlation was found between bacterial eggshell contamination with the
inoculated strain(s) on day 21 and shell penetration and whole egg contamination with the
strain(s). This corresponds with ample evidence in the literature that eggs with highly
contaminated eggshells suffer more from bacterial spoilage or whole egg contamination.
Messens et al. (2005) also showed that the probability of eggshell penetration was higher
with higher counts of Salmonella Enteritidis on the eggshell.
Using the agar approach Pseudomonas sp., Alcaligenes sp and Salmonella Enteritidis
penetrated most frequently. The higher shell contamination (on day 21) with Pseudomonas
sp. and Alcaligenes sp. can explain the higher fraction of penetrated eggshells.
Notwithstanding the comparable eggshell contamination with Salmonella Enteritidis and S.
warneri (both 2.5 log CFU/eggshell) on day 21, penetration prevalence with Salmonella was
higher (43% versus 18%). It is likely that the motile, non-clustering properties of Salmonella
favour the eggshell penetration.
Using the intact egg approach in the first study; Salmonella Enteritidis followed by
Carnobacterium sp. seemed to penetrate, survive and eventually grow most frequently;
respectively 33% and 17.5% of the inoculated eggs. Sauter and Petersen (1974) found a
contamination average of 47.5% for various salmonellae using whole eggs of poor shell
128
Aust. Poult. Sci. Symp. 2010...21
quality and 21.4% and 10.0% for whole eggs of intermediate and excellent shell quality.
Sauter and Petersen (1969) also challenged eggs with high, medium and low levels of shell
quality and found an incidence of spoilage with P. fluorescens of 6.3, 19.4 and 29.1%,
respectively. In our study 10.5% of the whole eggs were contaminated with Pseudomonas sp.
The second study showed no higher resistance of Salmonella enteritidis to egg
albumen compared to other salmonellae like Salmonella Typhimurium. Salmonella
Enteritidis strains originally isolated from the egg content were also not the primary invaders
of the egg content. Knowing that Salmonella Enteritidis is the most frequently isolated
Salmonella serovar in eggs, our results do not contra-indicate that the frequent egg content
contamination with Salmonella Enteritidis would be mainly due to the transovarian or vertical
route.
V. ACKNOWLEDGEMENT
This research would not have been possible without the technical help of especially Ann Van
de Walle. Sofie De Vlam, Elly Engels and Vera Van de Mergel are also gratefully
acknowledged.
REFERENCES
Alls AA, Cover MS, Benton WJ, Krauss WC (1964) Treatment of hatching eggs for disease
prevention - Factors affecting permeability and a visual inspection of drug absorption.
Avian Diseases 8, 245-246.
Barrow PA, Lovell MA (1991) Experimental infection of egg-laying hens with Salmonella
Enteritidis phage type 4. Avian Pathology 20, 335-348.
Berrang ME, Frank JF, Buhr RJ, Bailey JS, Cox NA, Mauldin J (1998) Eggshell
characteristics and penetration by Salmonella through the productive life of a broiler
breeder flock. Poultry Science 77, 1446-1450.
De Reu K, Grijspeerdt K, Messens W, Heyndrickx M, Uyttendaele M, Debevere J, Herman L
(2006) Eggshell factors influencing eggshell penetration and whole egg contamination
by different bacteria, including Salmonella Enteritidis. International Journal of Food
Microbiology 112, 253-260.
Drysdale EM (1985) Microbial penetration of the avian egg. PhD. Thesis. University of
Glasgow. Glasgow.
Himathongkham S, Riemann H, Ernst A. (1999) Efficacy of disinfection of shell eggs
externally contaminated with Salmonella enteritidis. Implications for egg testing.
International Journal of Food Microbiology 49, 161-167.
Humphrey TJ (1994) Contamination of egg shell and contents with Salmonella enteritidis: a
review. International Journal of Food Microbiology 21, 31-40.
Messens W, Grijspeerdt K, Herman L (2005) Eggshell characteristics and penetration by
Salmonella enterica serovar Enteritidis through the production period of a layer
flock. British Poultry Science 46, 694-700.
Nascimento VP, Cranstoun S, Solomon SE (1992) Relationship between shell structure and
movement of Salmonella enteritidis across the eggshell wall. British Poultry Science
33, 37-48.
Sauter EA, Petersen CF (1974) The effect of egg shell quality on penetration by various
salmonellae. Poultry Science 53, 2159-2162.
Sauter EA, Petersen CF (1969) The effect of egg shell quality on penetration by
Pseudomonas fluorescens. Poultry Science 45, 825-29.
129
Aust. Poult. Sci. Symp. 2010...21
THE EFFECTS OF TWO LIGHT-DARK SCHEDULES ON EGG LAYING TIME AND
SYNCHRONY, THE INCIDENCE OF LAYING IN THE DARK AND HEN WELFARE
G.M. CRONIN 1,2, S.S. BORG2, T.H. STOREY3, J.A. DOWNING2 and J.L. BARNETT 3
Summary
We investigated the effects of inserting a light period during the night on the shift of egg
laying times and inducing hens to lay in the dark, which would minimise egg laying at the
cage edges. Two methods of introducing light during the night were compared. A light period
was introduced either gradually, commencing with 0.5 h/night at 18 wks and increasing
incrementally to reach 3 h/night at 23 wks, or abruptly, by inserting 3 h/night at 23 wks. The
time and synchrony of egg laying, incidence of egg laying in the dark and stress response of
hens were measured. Regardless of the method of introducing the light period, median egg
laying time was shifted forward by about 1.5 h and about half the eggs were laid in darkness
(between 0300-0600 h). There were no effects on synchrony of laying or stress response,
measured via egg corticosterone. Photoperiod manipulation therefore offers a practical
method of reducing the high incidence of eggs laid in preferred sites such as the cage edges.
I. INTRODUCTION
The trend for modern layer cages to accommodate larger group sizes compared to earlier cage
designs, introduces potential egg quality problems if hens lay simultaneously in the preferred
sites in cages, which are the nest box (if present) and the edges of the cage (Cronin and
Barnett, 2008). Photoperiod strongly influences the timing of ovulation and oviposition in
laying hens (Morris, 2004). In environmentally controlled sheds with large group cages,
synchrony of egg laying may result in more eggs laid in preferred sites, increasing the
incidence of cracked shells as eggs roll into the collection tray and collide with other eggs.
Synchrony of egg laying may also result in increased stress on hens, if the preferred nest site
is not available due to coincidental occupation by cage-mates. A potential strategy to avoid
this problem is to decrease the proportion of hens that choose the ‘preferred’ egg laying sites.
Laying hens are inactive in the dark (Tanaka and Hurnik, 1991), including during the
pre-laying period if it coincides with darkness. Also, hens that lay in the dark tend to lay in
their current location, rather than in their ‘preferred’ egg laying site (Sherwin and Nicol, 1993).
Shifting oviposition time, so that some hens lay in the dark, offers a method of decreasing
competition for ‘preferred’ laying sites and reducing the risk of egg breakage. Manipulating
light-dark schedules, and in particular the time of actual lights-off and ‘expected sunset’ (viz.
the time birds expect darkness, which may be under the control of an internal time clock)
affect the time of ovulation and thus oviposition in hens (Lewis et al., 2007a and b). We
investigated the effects of two photoperiod schedules that involved introducing a period of
light to the birds during the night time, on the timing and synchrony of egg laying, incidence
of egg laying in the dark and hen welfare assessed by egg corticosterone concentrations.
II. MATERIALS AND METHODS
The experiment involved 96 Hy-Line Brown hens and was conducted in two adjoining
controlled environment rooms each containing six experimental cages. The hens were reared
1
Faculty of Veterinary Science, The University of Sydney, Camden NSW 2570 (present address)
Animal Welfare Science Centre, Department of Primary Industries, Werribee VIC 3030
3
Animal Welfare Science Centre, The University of Melbourne, Parkville VIC 3010
2
130
Aust. Poult. Sci. Symp. 2010...21
from day old to 13 weeks in group cages in a controlled environment shed. Pullets were not
beak trimmed and they were reared using conventional commercial lighting programs and
nutrition. At 13 weeks the birds were placed in groups of eight in Victorsson 8-bird cages
measuring 1.2 m wide and 0.5 m deep. All cages contained a nest box located on the right
side (viewed from the front), which measured 0.24 m wide and 0.5 m deep and the floor area
was covered with astro-turf. Nest box entrances were closed until 15 weeks. The same
photoperiod (i.e. total number of hours of light per day) was applied to both rooms,
increasing from 12 h at 15 weeks to 16 h at 23 weeks in 30 min increments weekly.
The experiment involved exposing birds to a period of light during the night, inserted
either gradually or abruptly. The treatments were assigned at random to one of two available
rooms. Thus, treatment was unavoidably confounded with room. Lighting was independently
controlled in each room and provided 20 lux at the level of the experimental cages. The
Gradual Light Introduction Treatment (Gradual) commenced at 18 weeks, when birds were
exposed to 30 min of light during the night (Table 1). The time of lights off for the inserted
light period was 0300 h daily and the duration of the light period was increased by 30 min per
week until 23 weeks of age, when the birds received 3 h of light at night within their overall
photoperiod of 16 h light to 8 h dark. The Abrupt Light Introduction Treatment (Abrupt)
commenced at 23 weeks, when birds were exposed to 3 h light inserted from midnight to
0300 h within their photoperiod of 16 h light to 8 h dark. Prior to 23 weeks however, the
Abrupt treatment served as a control for the Gradual treatment.
Low-light video cameras positioned below each cage and inside each nest box
provided a continuous view of the birds and enabled the time and location of egg laying to be
collated later from the digital video record (Cronin et al., 2007). While the number of eggs
laid in ‘preferred’ sites may be relevant to incidence of cracked shells, the present experiment
involved 8-bird cages as a model for cages with larger group sizes. Although the incidence of
cracked/broken eggs was recorded, the occurrence was insignificant and is not reported here.
All eggs laid each Friday were collected, numbered and taken to the laboratory where each
egg was broken to separate the albumen from the yolk. The albumen was weighed then
frozen for later analysis of corticosterone concentrations using the method developed by
Downing and Bryden (2008). Total corticosterone concentrations were assayed using a
commercial diagnostic kit (ICN ImmuChem Double antibody RIA, Seven Hills, NSW).
Table 1. Clock times (h) when the room lights were switched on and off each day over
progressive weeks of the experiment, and the median times of egg laying per week
for birds in the Gradual and Abrupt treatments.
Gradual treatment
Abrupt treatment
Median egg laying times
Hen age (wks) On Off On Off On Off On Off Gradual Abrupt SED
17
0600 1900
0600 1900
0740
0740 0.553
18†
0600 1900 0230 0300 0600 1930
0620
0730 0.601
19
0600 1900 0200 0300 0600 2000
0545 a 0740 b 0.473
20
0600 1900 0130 0300 0600 2030
0630 a 0730 b 0.316
21
0600 1900 0100 0300 0600 2100
0540 a 0720 b 0.298
22
0600 1900 0030 0300 0600 2130
0650 a 0830 b 0.511
23‡
0600 1900 0000 0300 0600 1900 0300 0600 0620 a 0805 b 0.620
24
0600 1900 0000 0300 0600 1900 0300 0600 0650
0725 0.659
25
0600 1900 0000 0300 0600 1900 0300 0600 0720
0715 0.554
26
0600 1900 0000 0300 0600 1900 0300 0600 0730
0735 0.625
Within hen age pairs, values with different superscripts differ a,b: P<0.05
SED: standard error of difference between the means
Start of Gradual and Abrupt Light Introduction Treatments indicated by † and ‡, respectively
131
Aust. Poult. Sci. Symp. 2010...21
Light schedule changes occurred on Monday and the number of eggs laid per hour
were collated from the video record of the next four days each week. Time of egg laying was
estimated as the median value per cage per 4-day period and synchrony was estimated as the
duration of the second plus third quartiles, that is, the time taken for the middle 50% of eggs
to be laid per day. Analysis of variance (GenStat 10.1, Lawes Agricultural Trust) was used to
test differences due to the treatments on egg-laying parameters and egg albumen
corticosterone concentrations and the experimental unit was the cage of birds.
III. RESULTS
Maximum hen day egg production was reached by both treatments at 20 weeks of age and
there was no effect of treatment on egg production. As shown in Table 1, the median egg
laying times in weeks 19-23 combined were about 1.5 h earlier each day in the Gradual
compared to Abrupt treatment. After 23 weeks, when both treatments received 3 h of
introduced light during the night, the median times did not differ. Although median egg
laying times were shifted earlier, there were no differences in the synchrony of egg laying,
which ranged from about 1.5-3.0 h. The introduction of light during the ‘normal’ night time
in the Gradual treatment, increased the proportion of eggs laid during darkness (Figure 1).
Between 18-23 weeks combined, 55.7% compared to 28.5% of eggs were laid in darkness in
the Gradual and Abrupt treatments, respectively (sed 16.98, P<0.01). In weeks 24-26
combined, following the introduction of the light period in the Abrupt treatment, there was no
difference in the proportion of eggs laid in darkness (38.4% and 51.2%, respectively for the
Gradual and Abrupt treatments, sed 8.87, P=0.18). Egg albumen corticosterone
concentrations did not differ between the treatments in weeks 18-23 combined (0.86 and 0.80
ng/g, sed 0.030) or weeks 24-26 (0.77 and 0.76 ng/g, sed 0.034).
Proportion of eggs laid (%)
100
Gradual
Gradual treatment - Light during night:
0h 0.5 h 1 h 1.5 h 2 h 2.5 h 3 h
Abrupt
80
60
40
20
Abrupt treatment commenced
3 h light during night
0
15
16
17
18
19
20
21
22
23
24
25
26
Age of hens (wks)
Figure 1. Eggs laid in the dark by birds in the Gradual and Abrupt Light Introduction treatments.
IV. DISCUSSION
Manipulation of the light schedule to insert of a period of light after midnight and before
0300 h, resulted in more eggs laid earlier in the day. Importantly, this coincided with a
planned period of darkness between 0300 and 0600 h. This was predicted, since the times of
132
Aust. Poult. Sci. Symp. 2010...21
lights on and lights off are two key factors influencing the time of ovulation and thus
oviposition in the laying hen (Lewis et al. 2007a). Neither the synchrony of egg laying nor
hen day egg production were altered by the treatments. Regardless of the light-dark schedule
manipulation, all birds were exposed to an increasing photoperiod, culminating in a total of
16 h of light per day by 23 weeks of age, which is the recommended photoperiod to ensure
good egg production in modern laying strains (Morris, 2004). Many studies are reported in
which manipulation of the light-dark cycle of laying hens has been investigated, including
comparisons of various intermittent light programs, including asymmetrical patterns,
symmetrical patterns with full light and symmetrical patterns with restricted light (van
Tienhoven and Ostrander, 1976; Morris and Bhatti, 1978; Morris and Butler, 1995; Morris,
2004). Provided birds received a minimum of 14 h light per 24 h, interruption of the night by
introducing a period of light did not affect shell breaking strength, egg size, hen day
production or hen mortality. Although we did not measure all these parameters in the present
experiment, as previously stated no egg quality problems were noted. An objective of this
research was to model the effects of light manipulation on egg laying patterns that could be
interpolated for use where larger cages housing larger groups of hens were used. The method
of achieving the light insertion during the dark period, that is whether it occurred gradually in
30 min increments per week over six weeks, or abruptly, produced a similar incidence of egg
laying in the dark and did not result in a stress response in the birds, measured via elevated
corticosterone concentrations in eggs.
In conclusion, the experiment demonstrated that a shift in egg laying time could be
achieved by inserting a period of light during the night. For about one-half of the hens,
oviposition could then coincide with darkness. One potential benefit of inducing a proportion
of birds to lay in the dark is that the number of eggs laid in ‘preferred’ egg-laying sites is
reduced, which could reduce the risk of cracked or broken eggs. No evidence of increased
stress response was detected, suggesting there were no adverse effects on bird welfare.
ACKNOWLEDGEMENTS
Financial support of AECL, DPI Victoria and the University of Sydney is gratefully
acknowledged. The experiment was conducted with the assistance of Bruce Schirmer who
cared for the birds and recorded and collected the eggs.
REFERENCES
Cronin GM, Barnett JL (2008) Final Report for Project VA-01 for the Australian Egg
Corporation Limited, AECL Publication.
Cronin GM, Borg SS, Fourdin SP, Storey TH, Barnett JL (2007) Proceedings, Australian
Poultry Science Symposium 19, 37-40.
Downing JA, Bryden WL (2008) Physiology and Behavior 95, 381-387.
Lewis PD, Ghebremariam WK, Gous RM (2007a) British Poultry Science 48, 239-244.
Lewis PD, Caston L, Leeson S (2007b) British Poultry Science 48, 276-283.
Morris TR (2004) World’s Poultry Science Journal 60, 163-175.
Morris TR, Bhatti BM (1978) British Poultry Science 19, 341-348.
Morris TR, Butler EA (1995) British Poultry Science 36, 531-535.
Sherwin CM, Nicol CJ (1993) Applied Animal Behaviour Science 36, 211-222.
Tanaka T, Hurnik JF (1991) Poultry Science 70, 483-488.
Van Tienhoven A, Ostrander CE (1976) Poultry Science 55, 1361-1364.
133
Aust. Poult. Sci. Symp. 2010...21
ATTRACTING LAYING HENS INTO RANGE AREAS USING SHELTERBELTS
E.A. BORLAND 1, S. HAZEL1, P.C. GLATZ 2, B.K. RODDA2, H. RIMMINGTON2, S.C.
WYATT2 and Z.H. MIAO2
In free range layer systems there is only a small percentage (9%) of birds that use the outdoor
range (Hegelund et al., 2005). This depends on the prevailing weather, season, age of birds,
flock size, time of day and the type of outdoor structures birds are provided (Hegelund et al.,
2005; Zeltner and Maurer, 2009). The birds that do range stay close to the house increasing
the stocking density (Hegelund et al., 2005) which may contribute to feather pecking.
A trial was undertaken to examine the role of shelterbelts in reducing feather pecking
behaviour by attracting hens into the range. A total of 120 Hyline Brown hens were housed at
17 weeks in an eco-shelter which had 6 internal pens (20 birds/pen) of equal size (2 m x 3 m)
and a free range area (726 m2) adjoining each pen. There were feeders, drinkers, nest boxes
and perches in each pen but no artificial light. Two treatments were examined; a shelterbelt
vs. no shelterbelt replicated 3 times. The treatment hens were provided a shelterbelt 10 m and
20 m from the poultry house while the control hens only had access to a bare paddock. The
shelterbelt consisted of a range of sizes of trees in pots [small (1 m high), medium (2 m high)
and large (3 m high)] as well as shrubs 1 m in height. Shelterbelts were established two
weeks before the trial commenced. From 24 to 31 weeks (May-July 2009) the birds were
allowed access to the range from 0800-1700 h. Measurements were made daily for egg
production, weekly for egg weight and fortnightly for feather score and body weight. On 5
occasions per week spot observations as well as weekly video recordings were made to assess
the behaviour of the birds in the range. The behavioural measures included the number and
distribution of birds <10 m from house and >10 m from house as well as feather pecking,
aggressive behaviours, dust-bathing, foraging, running, and comfort behaviours. Behavioural
recordings were analysed using Interact 8 software (Mangold, Germany).
There was a significantly (P = 0.019) higher percentage of birds in (67.4%) in the
range when provided a shelterbelt compared to 58.5% with no shelterbelt. In addition there
was a higher percentage (52.3%) of birds provided a shelter belt observed in areas >10m from
the house compared to the no shelterbelt (46.2%) treatment. Feather pecking was not
observed in either treatment throughout the experiment and other aggressive behaviours were
only observed once and therefore were not considered in the statistical analysis. Foraging and
running behaviour occurred significantly (P < 0.05) more in the shelterbelt treatment. There
was no significant effect on production and feather score of hens whether they were provided
shelterbelts or no shelterbelts in the range. The trial suggests that shelterbelts result in a
significant improvement in the percentage of birds using the range during the day.
Hegelund L, Sorensen JT, Kjaer JB, Kristensen IS (2005) British Poultry Science, 46, 1-8.
Zeltner E, Maurer V (2009) Proceedings, 8th European Symposium on Poultry Welfare. pp
104-112. Cervia, Italy. WPSA.
1
2
University of Adelaide, Roseworthy Campus, Roseworthy, SA 5371
SARDI, Pig and Poultry Production Institute, Roseworthy, SA 5371
134
Aust. Poult. Sci. Symp. 2010...21
ATTRACTING LAYING HENS INTO RANGE AREAS USING SHADE AND FORAGE
P.C. GLATZ 1, B.K. RODDA1, H. RIMMINGTON1, S.C. WYATT1 and Z.H. MIAO1
A concern for the free-range layer system is that only 9% of birds use the range area. The
factors which influence use of the range include weather (temperature, wind and rain),
season, age, flock size, time of day, shade and variety of overhead structures (Hegelund et al.,
2005, Zeltner and Maurer 2009).
Two trials were undertaken, one examined the role of shade areas in attracting laying
hens into the range and the other examined the role of forage. In the first trial a total of 120
laying hens (Hyline Brown) were housed at 18 weeks. The eco-shelter had 6 internal pens of
equal size (2m x 3m) with a free range area (726m2) adjoining the shelter. Hens were
provided feeders, drinkers, nest boxes and perches in each pen but no artificial light. There
were 2 treatments provided in the range, shade vs. no shade, with each treatment comprising
20 birds replicated 3 times. The control hens were not provided outdoor shade while the
treatment hens were provided a shaded area (3m x 2m x 1m= l x b x h) fitted with shade cloth
located 10 m and 20 m from the shed. From 32-44 weeks (March-May, 2008) hens were
allowed access to the range and measurements were made daily for egg production, weekly
for egg weight and four weekly for feather score. Video records were made of hens from each
of the replicates using the shade or in the range.
Shaded areas were visited by 18% of the hens with a tendency (P = 0.07) for more
hens to be in the paddock; 43% for paddocks with shade compared to 25% for the paddocks
with no shade provided. There was no significant effect on production and feather score of
hens whether they were provided shade or no shade in the range. The provision of shaded
areas in the free range attracted some additional hens into the range but other attractants are
needed.
In the second trial a total of 120 chickens (Hyline Brown) were randomly allocated
into 6 groups of 20 birds. There were 3 treatments provided in the range, no pasture (control);
vetch pasture and wheat pasture. Over the period 58-70 weeks of age (Aug–Oct 2008) birds
were allowed access to the range and measurements were made on production, feather score
and percentage of birds foraging. No difference in feather cover of the treatment groups were
observed but there was a significant interaction (treatment x time of day) for percentage of
birds foraging. The percentage of control birds in range was greater (55%) in morning than
the afternoon (30%) while for the pasture treatments about 45% if the birds were in the range
both during the morning and afternoon. The trial suggests that forage availability has an
impact on the percentage of birds using the range during the day.
Hegelund L, Sorensen JT, Kjaer JB, Kristensen IS (2005) British Poultry Science, 46, 1-8.
Zeltner E and Maurer V (2009) Proceedings, 8th European Symposium on Poultry Welfare.
pp. 104-112. Cervia, Italy. WPSA.
1
SARDI, Pig and Poultry Production Institute, Roseworthy, SA 5371
135
Aust. Poult. Sci. Symp. 2010...21
OPTIMUM LEVEL OF CASSAVA PULP IN DIETS FOR LAYERS
N. CHAUYNARONG 1, P.A. IJI1 and U. KANTO1
Summary
An experiment was conducted to determine the optimum level of cassava pulp that can be
included in layer diets without adverse effect on egg production and quality. Cassava pulp
replaced maize at 0, 5, 10, 15, 20 or 30 %, with equal metabolisable energy and crude protein
contents. The 30 % diet resulted in the lowest bulk density (610 g/litre). As the cassava pulp
level increased in the diet, the weight of the small intestine and bursa increased (P<0.001).
Inclusion of cassava pulp at 20 and 30 % in the diet was found to increase (P<0.001) feed
intake per dozen eggs and egg production was reduced (P<0.001) between 25 and 26 weeks
of age. Yolk colour score and shell thickness were adversely affected as cassava pulp level
was increased. It can be concluded that cassava pulp can be used at up to 15 % in layer diet
without detriment to egg production.
I. INTRODUCTION
Cassava is a cultivated crop that has been used widely as an energy source for livestock. The
principal characteristic of cassava flour is the high soluble carbohydrate (starch) compared to
other crops. It has been reported that cassava contains 40 and 25 % more carbohydrate than
rice and maize, respectively (Tonukari, 2004). Some studies have reported that whole cassava
can be used as a replacement of corn up to 60 % without adverse effects on egg production in
layers (Enriquez and Ross, 1972; Hamid and Jalaludin, 1972). Less research has been
conducted on cassava pulp, which is a residue obtained after extraction of starch from
cassava root (Sriroth et al., 2000). Chotineeranat et al. (2004) reported that cassava pulp still
contains up to 66 % starch. A large amount of this pulp is produced and is a potentially
inexpensive feed ingredient to replace corn, and other grains for livestock. However, there is
lack of information on its value as a feedstuff for poultry. Thus, the objective of this study
was to evaluate the potential value of pulp for laying hens, in terms of egg production and
egg quality.
II. MATERIAL AND METHOD
One thousand and two hundred Hisex layers were used in the experiment, between 16 and 28
weeks of age. The experiment was conducted in cages, each holding 2 birds. Twenty cages
(40 birds) constituted one replicate, with 5 replicates per treatment and there were 6
treatments in total, varying in cassava pulp content; 0, 5, 10, 15, 20 or 30 %. The cassava
pulp was obtained from Sonitch Starch Technology Co., Ltd, Thailand and the study was
conducted at a facility in Thailand. The diets were similar in energy and protein contents.
Birds had access to feed and water ad libitum. Temperature and ventilation were controlled
with an evaporative cooling system. Birds received a total of 17 hours of light per day. Room
temperature was controlled at 20-28 ˚C, and the relative humidity maintained between 60 and
70 %. Egg production and egg quality were assessed at 5 and 50 % lay, and at peak
production.
Ovary development was measured at 20 weeks of age; two birds per replicate were
randomly selected and slaughtered through cervical dislocation. The birds were dissected and
the weight of various visceral organs (ovary, proventriculus/gizzard, small intestine,
1
School of Environmental and Rural Science, University of New England Armidale NSW
136
Aust. Poult. Sci. Symp. 2010...21
pancreas, reproductive tract, spleen and bursa) and number of primary follicles were
recorded. Body weight was also recorded at the start and thereafter, every 4 weeks along with
feed intake. Data were analysed with Minitab statistical package. Results presented in this
study are mean values and differences between means were determined by the one-way
ANOVA.
III. RESULTS
(a) Bulk density
The bulk density of the diets was reduced as the level of cassava pulp in the diet was
increased in both phases (16-21 and 22-28 weeks of age). At 30 % cassava pulp in the diet,
bulk density was 610 grams/litre, compared to 775 and 750 g/litre for the control diets fed at
16-21 and 22-28 weeks, respectively.
(b) Ovary development and organ weight
The relative weight of the proventriculus and gizzard was found to be highest at 10 % cassava
pulp in the diet. Small intestinal weight increased (P<0.001) as cassava pulp level increased
in the diet. The weight of bursa also increased (P<0.05) in line with the level of cassava pulp
in the diet. However, the weight of other organs (ovary, reproductive tract, pancreas and
spleen) and number of primary follicles at 20 weeks of age were not affected by the level of
cassava pulp in the diets (Table 1).
Table 1.
Effect of cassava pulp level in the diet on ovary development at 20 weeks of
age (g/kg body weight)
Cassava Pulp
(g/kg)
NF
OW
P+G
SI
Pancreas
RT
Spleen
Bursa
0
1.08
6.7
29.7ab
49.3a
1.95
9.5
2.72
1.72a
50
0.22
2.4
29.0ab
52.5a
2.48
7.3
2.54
2.76b
100
0.26
3.0
32.0a
52.1a
2.29
7.7
2.43
2.36ab
150
0.33
3.0
27.4b
56.6ab
2.56
8.8
2.62
2.71b
200
0.40
2.2
26.1b
65.8bc
2.51
13.3
2.15
2.27ab
300
0.39
2.8
28.7ab
68.5c
2.07
9.1
2.61
2.88b
SEM
0.14
0.71
0.52
1.31
0.08
1.5
0.09
0.10
NF - Number of primary follicles, OW - Ovary weight (g) P+G - Proventriculus+gizzard weight (g),
RT - Reproductive tract weight (g). a,b,cMean values in the same column not sharing a
superscript are significantly different. SEM is standard error of difference between mean
values.
(c) Feed intake and body weight
Feed intake per day and feed intake per dozen eggs were not affected by level of cassava pulp
in the diet in the early lay period. However, as laying commenced inclusion of cassava pulp
at 20 and 30 % in the diet increased (P<0.001) feed intake per dozen eggs, while feed intake
was not affected at any level of cassava pulp in the diet. At the beginning, body weight was
not different between the treatment diets. However, at 20 weeks the hens fed with 30 % of
cassava pulp in the diet were found to be lighter and this was significantly different (P< 0.05)
from the control group. At 24 weeks of age, the lowest body weight was observed in hens on
the 30 % cassava pulp diet and this was significantly lower (P< 0.05) than on the 5 % diet.
137
Aust. Poult. Sci. Symp. 2010...21
However, at 28 weeks of age hens fed on the control (corn) diet had lower (P< 0.05) body
weight than the 5 % treatment group. At levels of cassava pulp higher than 5%, body weight
tended to be negatively affected.
(d) Egg production
Hen-day egg production during early lay (16-22 weeks) was not affected by level of cassava
pulp in the diet. At 23 and 24 weeks of age egg production tended to be reduced as the level
of cassava pulp was increased. The inclusion of cassava pulp in the diet at 20 and 30 %
significantly reduced (P<0.001) egg production between 25 and 26 weeks of age. However, at
27 and 28 weeks, the inclusion of cassava pulp at 5 to 15 % in the diet seemed to improve
egg production compared to the control diet (Figure 1). However, at 20% and 30 % of
cassava pulp egg production was reduced to the same level as on the control diet.
0
100
5
10
15
20
30
% Hen-day egg production
90
80
70
60
50
40
30
20
10
0
16
17
18
19
20
21
22
23
24
25
26
27
28
Age (w eeks)
Figure 1.
Effect of cassava pulp level in the diet on egg production at 16–28 weeks of age.
(e) Egg quality
At about 5 % lay, egg quality was not affected by any treatment diets. Yolk colour at 22 and
26 weeks of age was significantly reduced (P<0.001) with an increase in cassava pulp level in
the diet. Similarly shell thickness was reduced (P<0.05) at 22 weeks when the level of
cassava pulp increased to 30 %. However, shell thickness was better at 5% of inclusion than
on the control diet.
IV. DISSCUSION
The results showed that feed intake per dozen eggs was increased at 20 and 30 % of cassava
pulp in the diets. This may be due to the bulk density of the feed. Feed flow could be reduced
at lower bulk density, and thus affect feed consumption and egg production (Fairfield, 2003).
The inclusion of cassava pulp at 5-15% in the diet seemed to improve egg production when
compared to the control diet. This could be related to the increase in weight of small intestine
138
Aust. Poult. Sci. Symp. 2010...21
as the cassava pulp level was increased. In the experiment of Nestor et al. (1981), the weight
of the small intestine was affected by high-fibre diet but this had no effect on egg production
or egg weight. The reduction in yolk color with increase in the level of cassava pulp may be
due to the absence of pigments in the ingredient. Shell thickness was also found to decline
with cassava pulp level, in line with the report by Tani et al. (2005) that high fibre reduces the
absorption of calcium. It can be concluded that cassava pulp can be used as an alternative
ingredient to maize at up to 15 % without negative effects on egg production. However,
artificial pigments might be needed to maintain yolk colour. Further studies are planned to
examine the benefits of supplementation with microbial enzyme.
REFERENCES
Enriquez FQ, Ross E (1972) Poultry Science 51, 228-232.
Hamid K, Jalaludin S (1972) Malaysian Agricultural Research 3, 48-53.
Chotineeranat S, Pradistsuwana C, Siritheerasas, C, Tantratian S (2004) Journal of Science
Research Chulalongkorn University 29, 119-127.
Sriroth K, Chollakup R, Chotineeranat S, Piyachonkwan K, Oates CG (2000) Processing,
Bioresource Technology 71, 3-69.
Bertol TM, Lima GJMM de (1999) Pesquisa Agropecuaria Brasileira 34, 243-248.
Fairfield DA (2003). Feed and Feeding Digest 54, 1-5.
Nestor KE, Cantor AH, Bacon WL, Brown KI (1981) Poultry Science 60, 1458–1467.
Tani M, Ikeda R, Shinjo K, Kobayashi K (2005) Jin'ai Joshi Tanki Daigaku Kenkyu Kiyo 37,
37-46.
139
Aust. Poult. Sci. Symp. 2010...21
IMPLICATIONS ON THE USE OF CANOLA MEAL AND RECOMMENDATIONS TO
REDUCE FISHY TAINT IN EGGS FROM BROWN HENS
R.A. PEREZ-MALDONADO 1 and T.TRELOAR
Canola meal (CM) is a source of highly digestible protein (360-420 g/kg DM) with low antinutritive factors making it an ideal feed ingredient for intensive livestock. But, the use of
high levels of CM in brown layer diets leads to the production of a “fishy” taint in eggs. The
smell is caused by the presence of trimethylamine (TMA) which is generated by the high
level of sinapine (10-15 g/kg DM) found in Australian CM. Sinapine passes to the gut and is
converted to choline and further metabolised to TMA. In white layers TMA is effectively
metabolised and excreted. But a significant percentage of brown hens have low enzyme
levels and when the amount of TMA exceeds the capacity of the hen to metabolise and
excrete it, some TMA is diverted to the developing ova, producing a “fishy” taint in the egg.
The effects of sinapine, choline, and CM on the generation of fishy taint were studied
using populations of brown layers known to lay taint affected eggs and layers that produced
eggs without fishy taint (non-responders). The amount of supplemented choline in the diet
significantly increased the occurrence and severity of the fishy taint when the total level of
choline in the diet reached 2600 mg/kg (twice the daily recommended allowance). The effect
of CM inclusion with known endogenous sinapine content, show that a small amount of taint
was present in all of the hens groups, including the group that was fed zero CM and also the
group of non-responders hens that was fed a maximum inclusion level of 15%. There was no
significant difference in either the occurrence or severity of taint up to an inclusion of 12%
CM (containing 1.46 g/kg endogenous sinapine). Increasing the level of CM to 15%
(containing 1.83 g/kg endogenous sinapine) led to a significant increase in the level of taint
(see Table).
Treatment
A 0% CM
B 6% CM
C 9% CM
D 12% CM
E 15% CM
F 15% CM*
Sinapine
(g/kg)
trace
Choline
(mg/kg)
1082
0.73
1236
1.10
1355
1.46
1490
1.83
1595
1.83
1595
Set 1
Affected
Eggs
Set 2
Affected
Eggs
Set 3
Affected
Eggs
Overall Affected
Eggs (%)
1/10
2/10
1/10
13.3%
1/10
1/10
2/10
13.3%
1/10
1/9
2/10
13.8%
2/10
2/10
1/10
16.7%
3/10
4/9
3/9
35.7%
1/10
0/10
0/9
3.4%
*Note: Treatment F contain layers not enzyme deficient, effectively degrading TMA
The total endogenous choline level in the diets increased from 1082 mg/kg (diet
without CM), to 1595 mg/kg in the diet with 15% CM. While our work have shown that this
level of choline (when using supplemental choline chloride) did not result in an increase in
taint by itself, it is possible that such an increase in the amount of choline (endogenous
choline from CM) may have influenced the observed levels of taint in the presence of
sinapine. As CM is high in both, choline and sinapine (endogenous source), the combined
effect of these two compounds on the generation of taint is an important factor in the
usefulness of CM in layer diets and requires more attention.
1
Rider Perez-Maldonado Consultancy, 36 Forest Lake Qld 4078
140
Aust. Poult. Sci. Symp. 2010...21
EFFECT OF A COMPLEX ENZYME ON THE DIGESTIBILITY OF LAYER DIETS
CONTAINING PALM KERNEL MEAL
V. RAVINDRAN 1, D.THOMAS1, A. LEARY 2 and A. KOCHER 3
Summary
The effects of a broad range activity enzyme complex produced by solid state fermentation
(Allzyme SSF) was tested using a commercial peak layer diet and an experimental diet containing
palm kernel meal (25% w/w substitution). The addition of a solid state fermentation enzyme
resulted in improvements in energy in all diets. In addition diets containing 25% PKM plus
enzyme had a significantly better fat digestibility as well as numerically improved crude
protein and amino acid digestibility. This study demonstrated that the use of a broad range
activity enzyme complex produced by solid state fermentation leads to increased energy and
nutrient release in laying hen diets with increased levels of palm kernel meal.
I. INTRODUCTION
Last year, feed prices reached record highs due to the global demand of corn and soyabean
and partly because of climatic conditions and poor harvests in same parts of the world. Many
producers faced a huge challenge to balance nutrient requirements for their animals and the
cost of diets without negatively impacting on overall animal performance. The use of byproducts from the oilseed industry (rapeseed, coconut or palm kernel) and the use of
endogenous feed enzymes have become paramount in maintaining poultry performance at
affordable levels.
The two largest producer of palm nuts and palm kernel meal (PKM) or palm kernel
cake (PKC) are Malaysia and Indonesia (FAOSTAT, 2007). The rising cost of transport and
the cost of conventional feed material make PKM a product of increasing importance for the
local poultry industry in Asia. PKM is the residue of the palm kernel after the solvent
extraction of oil. The nutritive value will depend on the species of palm nut, the oil extraction
process and the shell residue in the meal. Although PKM is considered a to be a reasonable
source of protein for poultry, the inclusion level is limited due to its high content of insoluble
NSP and its low amino acid content (Sundu et al., 2006). The majority of non-starch
polysaccharides (NSP) are in the form of insoluble linear mannan (78%) with low
substitution of galactose (Düsterhöft et al., 1992). The lack of endogenous enzymes to
depolymerise these mannans manifests itself in poor growth and feed utilisation when PKM
is included at levels above 10% in broiler diets or layer diets (Chong et al. 2008; Soltan,
2009). On the other hand, reports by Perez et al. (2000) and Sundu et al. (2005) indicated that
bird performance was not affected with 40% of PKM in the diet. It has to be noted that in
both these studies the diets were balanced for amino acids. The availability of amino acids in
PKM is generally high; however, levels of lysine and methionine are significantly lower in
comparison to soyabean meal. Furthermore, PKM has a wide ratio (3.7 to 3.9) of argentine to
lysine (Sundu et al., 2006), since the nutritional requirement of lysine, methionine and
argentine are interrelated, diets with high levels of PKM need to be balanced for these three
amino acids (Chamruspollert et al., 2002).
1
Institute of Food, Nutrition and Human Health. Massey University, NZ
Alltech Bangkok, Thailand.
3
Alltech Biotechnology Pty Ltd. Dandenong South, Vic
2
141
Aust. Poult. Sci. Symp. 2010...21
A number of studies have shown that the use of endogenous feed enzymes can
improve digestibility of diets with levels of PKM above 20% (Chong et al., 2008; Sundu et
al., 2005). In these studies the actual bird performance was not significantly improved in
enzyme supplemented diets. Many commercially available feed enzymes are a focusing on
the depolymerisation of the insoluble carbohydrates in PKM, little attention has been given to
the changes in amino acid digestibility. This paper reports the result of a study investigating
the effect of an exogenous enzyme complex on the apparent metabolisable energy and ileal
amino acid digestibility of laying hen diets.
II. MATERIALS AND METHODS
The effects of a commercially available feed enzyme (Allzyme SSF, Alltech Inc, USA;
included at 150g/t) was tested using a commercial peak layer diets (Mainland Poultry, NZ)
and an experimental diet containing palm kernel meal (25% w/w substitution). Titanium
oxide was added to all diets as a digesta marker (3 g/kg). A total of 64 layers (Hi Line; 30week old) of uniform body weight, were allocated to individual cages housed in an
environmentally-controlled shed. Four diets were fed for 10days. During the last 4 days, feed
intake and excreta output were measured quantitatively per cage for the determination of
AME. On the last day, all birds were euthanased by an intracardial injection and the contents
of the lower half of the ileum were collected. Diet and ileal digesta samples were analysed for
dry matter, titanium, fat, protein and amino acids, including methionine and cystine, but not
tryptophan.
III. RESULTS
Table 1.
a,b
Influence of an enzyme complex on the apparent metabolisable energy and
ileal digestibility coefficients of fat and protein
AME MJ/kg DM
Fat
Crude protein
Control
Control + enzyme.
Probability
Pooled SEM
12.99b
13.14a
0.05
0.039
0.864
0.901
0.16
0.016
0.780
0.811
0.07
0.010
PKM diet
PKM diet+enzyme.
Probability
Pooled SEM
11.41b
11.64a
0.001
0.023
0.893x
0.928y
0.05
0.007
0.650
0.693
0.12
0.017
PKM
PKM + enzyme
Probability
Pooled SEM
6.69b
7.13a
0.05
0.091
0.980
1.008
0.53
0.029
0.259
0.342
0.42
0.068
Means bearing different superscript in a row are statistically different (P<0.05).
The addition of enzyme resulted in small, but significant (P < 0.05), improvements in the
AME of the control diet (Table 1). Numerical increases were seen in the ileal digestibility of
fat (P > 0.05) and protein (P = 0.07) when enzyme was added to the diet. Furthermore,
addition of enzyme resulted in significant (P < 0.05) improvements in the digestibility of
alanine and histidine, and tended (P < 0.10) to improve those of proline, valine and leucine
(Table 2). The addition of the enzyme complex significantly increased (P < 0.05) the AME of
142
Aust. Poult. Sci. Symp. 2010...21
PKM diet by 0.23 MJ/kg. Ileal fat digestibility was significantly improved (P<0.05) by the
enzyme. A 4% unit improvement in ileal protein digestibility was observed with the addition
of the enzyme complex, but this difference was not significant (P = 0.12). Digestibility of
several amino acids tended to be increased with the addition of SSF and these included
aspartic acid (P = 0.07), threonine (P = 0.11), serine (P = 0.07), glutamic acid (P = 0.08),
alanine (P = 0.09), valine (P = 0.08), leucine (P = 0.08), phenylalanine (P = 0.07),histidine (P
= 0.06), lysine (P = 0.06) and arginine (P = 0.08). The average digestibility of amino acids
tended (P = 0.08) to be greater when enzyme was added to the diet. In calculating the values
for the AME and ileal digestibility of fat, protein and amino acids in PKM and the effects of
enzyme it was assumed that there were no interactions between the basal diet, PKM and SSF.
The values calculated for PKM were found to be very low – especially ileal digestibility
coefficients of protein and amino acids. It is likely that the assumption of no interaction may
not have been applicable for a high-fibre ingredient such as PKM and this may be responsible
for the apparently spurious results. The only notable observation is that the calculated AME
of PKM was increased (P<0.05) from 6.69 to 7.13 MJ/kg – an increase of 0.44 MJ/kg (105
kcal/kg).
Table 2.
Treatment
Influence of an enzyme complex on the ileal digestibility of amino acids
Lys.
Met.
Arg.
Pro.
Ala.
b
Val.
Leu.
His.
b
Mean
Control
Cont+enzyme.
Probability
Pooled SEM
0.870
0.881
0.46
0.010
0.830
0.831
0.95
0.015
0.834
0.846
0.52
0.013
0.771
0.810
0.08
0.013
0.842
0.888a
0.05
0.012
0.813
0.843
0.09
0.011
0.856
0.881
0.08
0.009
0.815
0.857a
0.05
0.012
0.798
0.823
0.16
0.011
PKM diet
PKMdiet+enz.
Probability
Pooled SEM
0.752
0.807
0.06
0.017
0.731
0.758
0.33
0.018
0.718
0.762
0.08
0.15
0.633
0.651
0.48
0.016
0.753
0.791
0.09
0.013
0.691
0.731
0.08
0.013
0.761
0.798
0.08
0.012
0.633
0.690
0.06
0.017
0.669
0.710
0.08
0.013
PKM
PKM+enzyme
Probability
Pooled SEM
0.396
0.583
0.10
0.068
0.436
0.540
0.35
0.072
0.369
0.511
0.15
0.060
0.221
0.173
0.62
0.065
0.487
0.497
0.38
0.053
0.323
0.396
0.38
0.054
0.475
0.549
0.33
0.049
0.088
0.188
0.35
0.069
0.282
0.372
0.28
0.054
a,b
Means bearing different superscript in a row are statistically different (P<0.05). Average is the
mean of 17 amino acids.
IV. DISCUSSION and CONCLUSIONS
The use of exogenous feed enzymes is a common tool in the poultry industry to increase
nutrient availability in cereal based diet. The addition of a multi activity enzyme complex to
either a commercial diet or a diet with 25% PKM resulted in a significant increase in AME of
1.1% in the control diet 2% in a diet with 25% PKM or 6.66% on the basis of PKM alone.
The higher AME values in the present of the same enzyme preparation are consistent with
previous work, (Wu et al., 2004) reported improvements between 1.9% (sorghum), 2.1%
(wheat), 2.6% (maize) and 7.8% (barley) in the presence of enzyme and (Chong et al., 2008)
showed a 6.0% improvement in TMEn in diets containing 25% palm kernel cake.
The improvement in nutrient digestibility in the presence of exogenous feed enzymes
is the result of depolymerising soluble NSP causing high intestinal viscosity and reduced
nutrient digestibility but it also can be a direct result of the enzymatic breakdown of the cell
143
Aust. Poult. Sci. Symp. 2010...21
walls and the release of entrapped nutrients (Choct, 2006). Industrial enzymes used in the
animal feed industry are produced either by submerged liquid fermentation or by solid
substrate fermentation (SSF). Enzyme production using SSF technology, which is
characterised by growing a selected strain of yeast, bacteria or fungi on a water insoluble
substrate, results in a wider range of enzymatic activities which are not produced by
submerged liquid fermentation (Filer, 2008).
The addition of a solid state fermentation enzyme resulted in improvements in energy
in all diets, in addition diets containing 25% PKM plus the enzyme complex had a
significantly better fat digestibility as well as numerically improved crude protein and amino
acid digestibility. It is likely that the board range of enzyme activities produced in the solid
state fermentation process was able to release some of the nutrients entrapped in the
rectangular honeycomb-like cell wall of the palm kernel (Chong, 1999) resulting in increased
digestibility of nutrients. This study demonstrated that the use of an broad range activity
enzyme complex produced by solid state fermentation lead to increased energy and nutrient
release in laying hen diets with increased levels of palm kernel meal.
REFERENCES
Chamruspollert M, Pesti GM, Bakalli RI (2002) British Journal of Nutrition 88, 655-660.
Choct M (2006) World's Poultry Science Journal 62, 5-16.
Chong CH (1999) University of British Colombia.
Chong CH, Zulkifli I, Blair R (2008) Asian-Australasian Journal of Animal Science 21,
1053-1058.
Düsterhöft EM, Posthumus MA, Voragen AGJ (1992) Journal of the Science of Food and
Agriculture 59, 151-160.
FAOSTAT (2007). In: Food and Agriculture Organisation of the United Nations.
Filer K (2008). In 'Feed Tech'. pp. 16-17.
Perez JF, Gernat AG, Murillo JG (2000) Poultry Science 79, 77-79.
Soltan MA (2009) Livestock Research for Rural Development 21.
Sundu B, Dingle J, Kumar A (2006) World Poultry Science Journal 62, 316-325
Sundu B, Kumar A, Dingle J (2005) Proceedings, Australian Poultry Science Symposium
17, 227-228
Wu YB, Ravindran V, Hendriks WH (2004) Journal of the Science of Food and Agriculture
84, 1817-1822.
144
Aust. Poult. Sci. Symp. 2010...21
OPPORTUNITIES AND SUSTAINABILITY OF SMALLHOLDER POULTRY
PRODUCTION IN THE SOUTH PACIFIC REGION
P.C. GLATZ 1, I.D. BLACK 2, W. AYALEW 3, J.K. PANDI3, R.J. HUGHES1, Z.H. MIAO1, J.
WAHANUI 4, T. JANSEN 5, V. MANU 6 and B.K. RODDA1
Summary
Village poultry provides a source of income, improves human nutrition and helps meet family
and social obligations. Local feed resources are available in the Pacific that could be utilized
more effectively for feeding poultry. Balanced rations for village birds can be devised based
on the variety of potential feeds. An education program based on new feeding systems can
then be extended to farmers. New crops and pasture species with higher nutritional value for
poultry could also be introduced. Where smallholders use imported commercial feeds the
viability of these operations has been threatened by rising costs of imported feeds. To
overcome this problem there is potential to use a concentrate diet that can be blended with
cheap local feed ingredients or to dilute a commercial diet with cheap locally available
ingredients with little impact on production and sale of birds and eggs. Alternatively mini
mills could be established to make diets in areas where appropriate local feed ingredients are
available and cost competitive. Adoption of such feeding systems is considered a solution for
the viability of village broiler farming in Papua New Guinea (PNG) and is a potential method
of supporting smallholder poultry operations in other areas of the Pacific.
I. INTRODUCTION
Smallholder poultry farming makes an important contribution to the livelihoods of rural
households in the Pacific. Under the traditional “village bird” system on smallholder farms
poultry are allowed to scavenge for feed, have little shelter and are free to wander around the
village. This system is cheap and the farmers have little work to do. There are minimal
provisions for shelter, feed and water and few management skills are required. However there
are considerable problems with this system including slow growth and poor productivity due
to energy and protein deficiencies, poor bird genetics, losses due to predators and theft and
damage to village gardens. However, many smallholders have changed from producing
poultry solely for household purposes to producing meat birds and eggs for sale in local
markets. They have adopted improved housing and nutrition and use modern genetic strains
fed on commercial feeds. However the viability of these semi commercial village operations
has been threatened by the rising costs of imported ingredients and feeds. In Pacific countries
this problem is primarily due to the reduction in the value of the local currencies and
consequent increases in the cost of all imports. This has contributed to the demise of some
smallholder operations relying on imported feed. When transport problems are added (and
these are also made worse by the rising real cost of fuel), the smallholder poultry industries
that rely on imported ingredients are in difficulty. The viability of the smallholder operations
1
SARDI, Pig and Poultry Production Institute, Roseworthy, SA 5371
SARDI, Waite Plant Research Centre, GPO Box 397, Adelaide, SA 5001
3
NARI, PO Box 1639, Lae 411, Morobe Province, Papua New Guinea.
4
DAL, PO Box G13, Honiara, Solomon Islands.
5
TerraCircle Association Inc, 315 Afterlee Road, Kyogle, NSW 2474
6
MAFF, PO Box 14, Nuku’alofa, Tonga.
2
145
Aust. Poult. Sci. Symp. 2010...21
will be substantially improved if feeding regimes based at least in part on local ingredients
can be developed as an alternative to imported complete feed or feed ingredients from
overseas.
II. OPPORTUNITIES
Local feed resources are available in the Pacific that could be utilized more effectively for
feeding poultry (ALFID, 2002). Farmers could also introduce new crops and use pasture
species with higher nutritional value for poultry (Glatz, 2008). Effective rations for village
poultry can be developed based on feed resources available. Four feeding strategies for
poultry could be adopted by smallholder farmers; a) complete ration formulation using local
feed ingredients; b) free choice of feed ingredients; c) mixing a concentrated diet with local
feed ingredients and d) dilution of a commercial diet with locally available food products.
(a) Development of village poultry rations for village farmers in the Solomon Islands
There is a wide variety of local feed resources available that could be utilized more
effectively such as root crops, fruit, forages, bush plants and vines. Farmers in remote areas
can introduce new crops (sorghum, mung bean, pigeon pea, sunflower, amaranth and others)
with higher nutritional value for poultry and provided they have the appropriate tools and
growing technologies. For example in the Solomon Islands three feeding trials with village
hens were conducted using local feed resources.
Composition of diet 1: Corn (45%), fresh grated cassava (6.31%), ripe paw paw (5%),
mung beans (30%), fishmeal (5%), lime (8%), premix 0.25%, lysine 0.09%, methionine
0.05% and salt (0.3%). Specifications of diet; AME 10.8 MJ/kg; CP 147 g/kg; fat 26 g/kg;
fibre 22g/kg, Ca 30 g/kg and P 5 g/kg (calculated values).
Composition of diet 2: Corn (25%), pigeon pea (15%), ripe paw paw (5%), mung
beans (30%), fresh grated coconut (5.81%), fresh grated cassava (10%), lime (8%), dicalcium
phosphate (0.5%), premix 0.25%, lysine 0.09%, methionine 0.05% and salt (0.3%).
Specifications of diet; AME 9.3 MJ/kg; CP 136 g/kg; fat 33 g/kg; fibre 30g/kg, Ca 30 g/kg
and P 5.2 g/kg (calculated values).
Composition of diet 3: Pigeon pea (25), sorghum (45%), ripe paw paw (8%), fresh
grated cassava (8.31%), fishmeal (5%), lime (8%), premix 0.25%, lysine 0.09%, methionine
0.05% and salt (0.3%). Specifications of diet; AME 10.2 MJ/kg; CP 125 g/kg; fat 27 g/kg;
fibre 36g/kg, Ca 30 g/kg and P 5.5 g/kg (calculated values).
The trials compared the performance of birds fed the local home mix layer ration with
an imported commercial layer feed. The feeding trials were conducted in a naturally
ventilated barn at the Solomon Islands College of Higher Education comprising 16 pens (1.5
x 1.5m) each with perches, nest boxes, drinkers and feeders. A total of 64 local hens were
used for each trial which included 4 replicates of a control commercial layer diet (as the gold
standard) and 4 replicates of the local feed diet. The birds were of a mixed age and obtained
from local farmers. Corn and mung beans were included as whole grain in the rations.
Cassava and paw paw were chopped, weighed, mixed and fed fresh twice daily. Numbers of
eggs laid were recorded daily. Egg production was significantly lower in birds fed the local
mix ration compared to the commercial ration (Figure 1). However the cost of using the
imported commercial feed is too expensive for village farmers to consider purchasing (Glatz,
unpublished).
146
Aust. Poult. Sci. Symp. 2010...21
60
Egg Production %
50
40
Control
Diet 1
30
Diet 2
Diet 3
20
10
0
0
2
4
6
8
10
Week
Figure 1.
Egg production % of village hens in the Solomon Islands fed a commercial
ration vs. three local diets.
In Tonga the composition of the local ration was maize (45%), cassava meal (15%),
copra meal (16.56%), fish meal (15%), crushed sea shells (7%), salt (0.3%), dicalcium
phosphate (1%), lysine (0.09%), methionine (0.05%). Specifications of diet; AME 12.2
MJ/kg; CP 181 g/kg; fat 53 g/kg; fibre 7 g/kg, Ca 32 g/kg and P 6.8 g/kg (calculated values).
A total of 36 commercial hens (34-37 weeks) were used for the poultry trial, with two
birds/cage (30cm wide x 45 high and 30 cm deep) located in the middle row of naturally
ventilated single tier cage shed. The diets tested were; 1) commercial imported layer diet, 2)
commercial imported layer diet diluted with 30% copra meal and the local ration. A
commercial layer diet diluted with 30% of copra meal resulted in similar egg production
compared with the imported commercial layer diet. However, layers fed on the diet
formulated with local feed produced the lowest (P<0.05) numbers of eggs. This was due to
the batch of copra meal that was added to the local diet being rancid.
Table 1.
Egg production % of birds in Tonga fed a commercial ration, a commercial
ration diluted with 30% copra meal and the local diet.
Treatment
Production (%)
Commercial
94.3a
70% Commercial+30% copra meal
95.7a
Local feed
27.1b
P
0.026
Means with a common letter are not significantly different (P > 0.05)
147
Aust. Poult. Sci. Symp. 2010...21
(b) Free choice feeding in village poultry
In village layers, Grayson and Campbell (2004) reported that free choice feeding system can
be used successfully for village poultry in the Solomon Islands. Feed from the three food
groups; protein, carbohydrate and minerals and vitamins are provided separately every day so
the birds can choose feed according to their requirements.
(c) Development of a concentrated diet that can be blended with local feed ingredients
Glatz (2006) reported on a feeding strategy whereby PNG fishmeal and copra meal (plus
minerals and vitamins) were used to produce a concentrate that can be supplemented with 5080% of local ingredients (e.g. sweet potato) to make up the whole ration. The promising
results from an on-station evaluation of the concentrate were followed by an on-farm
evaluation in the Eastern Highlands Province in PNG. The trial with 60 smallholder village
farms (each growing 50 meat birds) compared the performance of birds fed a commercial diet
vs. a diet comprising the poultry concentrate fed with cooked sweet potato. While the higher
moisture content of the treatment diet (containing 70% sweet potato) resulted in a higher feed
intake (4.26 vs. 4.0 kg/bird) and a lower live sale weight (2.0 vs. 2.4 kg), the income from
sale of live birds was only about 5% lower for the concentrate plus sweet potato diet.
Pandi and Ayalew (unpublished) subsequently tested four diets in a large broiler grow
out trial at the National Agriculture Research Institute (NARI) in Lae. The diets were; 1) 50%
sweet potato plus 50% low energy concentrate; 2) 70% sweet potato plus 30% low energy
concentrate; 3) 50% cassava plus 50% high energy concentrate and 4) 70% cassava plus 30%
low energy concentrate.
The composition and calculated nutrient specifications of high energy and low energy
concentrate were as follows:
(a) High energy concentrate (11.6MJ/kg); Composition (g/kg), sorghum 118; soya 485.9;
meat & bone meal 286; tallow 65; L-lysine 12.5; DL-methionine 11.8; L-threonine 1.3; salt
5; mycocurb 1; choline cloride 4.5; boiler premix 9; Specification (g/kg); Protein (419.75),
Arg (28.59), Isoleu (15.95), Lys (32.89), Meth (17.41), M+C (23.25), Threo (16.58), Fibre
(23.62), Ca (27.56), Ptot (16.73), AvP (9.4), NA (3.58), K (10.48), CI (6.08)
(b)Low energy concentrate (9.4MJ/kg); Composition (g/kg) mill run 246; soya 389.45; meat
& bone meal 309; + tallow 8; L-lysine 13.8; DL-methionine 12.4; L-threonine 1.7; salt 5;
mycocurb 1; choline chloride 4.5; ronozyme phytase 0.15; broiler premix 9; Specification
(g/kg); Protein (417.87), Arg (28.19), Isoleu (14.91), Lys (33.01), Meth (18.01), M+C
(24.03), Threo (16.45), Fibre (38.47), Ca (30.29), P tot (20.18), AvP (11.44), Na (3.7), K
(10.5), CI (6.43).
Each diet was replicated 4 times in 16 floor pens in a naturally ventilated shed.
Results indicated favorable bird performance with all the feeding options assessed. However,
the better performing feeds when assessed for overall weight gain, feed intake and feed
conversion were; 50% sweet potato plus 50% low energy concentrate; 70%; sweet potato
plus 30% low energy concentrate and 50% cassava plus 50% high energy concentrate (Table
3).
148
Aust. Poult. Sci. Symp. 2010...21
Table 3.
Age
(weeks)
Weekly body weights (kg) of meat chickens grown at NARI on combinations
of concentrate, sweet potato and cassava compared to a commercial grower
ration.
Commercial
Grower
SP50
LEC
C50
HEC
SP70
LEC
C70
HEC
LSD
P value
3
0.90a
0.91a
0.91a
0.90a
0.91a
0.012
0.838
4
1.27a
1.23b
1.24ab
1.08c
1.09c
0.041
<.001
5
1.86a
1.59b
1.53c
1.49c
1.24d
0.055
<.001
Means with a common letter within a row are not significantly different (P > 0.05). SP50=50% sweet
potato; SP70=70% sweet potato; C50=50% Cassava; LEC=low energy concentrate; HEC=high
energy concentrate
The two most promising feeding options that were tested based on the results in Table
3 were diets comprising 50% cassava mixed with a high energy concentrate and 50% sweet
potato mixed with a low energy concentrate. It was considered that the best bet feeding
options that required further evaluation in lowland and highland areas of PNG were 50%
sweet potato and 50% LEC; 50% Cassava and 50% HEC and 70% sweet potato and 30%
LEC. These diets were tested at 3 different sites in PNG; Christian Leaders Training Centre
(CLTC), Lutheran Development Service (LDS) and OK Tedi Development Foundation
(OTDF). The grow out facilities at each of these sites comprised 8 floor pens in a naturally
ventilated shed with two replicates of each diet.
Table 4a.
CLTC site; weekly body weights of meat chickens fed on combinations of
concentrate, sweet potato and cassava.
Experimental Diet
Day 21
Day 28
Day 35
Day 42
0.810
0.813
0.813
0.810
1.371a
1.276a
1.221a
0.906b
1.988a
1.735b
1.566b
1.157c
2.724a
2.283b
2.082c
1.514d
P-value
0.803
0.004
<0.001
LSD
0.009
0.147
0.174
Means with a common letter within a row are not significantly different (P > 0.05)
<0.001
0.151
Commercial finisher
50SP + 50 LEC
50C + 50 HEC
70SP + 30 LEC
Table 4b.
LDS site; weekly body weights of meat chickens fed on combinations of
concentrate, sweet potato and cassava.
Experimental diets
Day 21
Day 28
Day 35
Day 42
Commercial finisher
50SP + 50 LEC
50C + 50 HEC
70SP + 30 LEC
0.801
0.800
0.801
0.801
1.477a
1.318b
1.327b
1.154c
2.011a
1.802b
1.860b
1.454c
2.724a
2.342b
2.377b
1.785c
P-value
LSD
0.651
0.002
<0.001
0.067
<0.001
0.099
<0.001
0.200
149
Aust. Poult. Sci. Symp. 2010...21
Table 4c.
OTDF site; weekly body weights of meat chickens fed on combinations of a
concentrate, sweet potato potato and cassava.
Experimental diets
Day 21
Day 28
Day 35
Day 42
Commercial finisher
50SP + 50 LEC
50C + 50 HEC
70SP + 30 LEC
0.870
0.878
0.870
0.880
1.536a
1.365b
1.339b
1.174c
2.181a
1.916b
1.919b
1.627c
2.570a
2.463a
2.457a
2.010b
P-value
LSD
0.598
0.024
0.012
0.149
<.001
0.111
0.008
0.224
The results (Tables 4a, 4b, 4c) from the three climatic zones in PNG confirmed
findings in grow-out trials at NARI that birds fed the 50% sweet potato with 50% low energy
concentrate and 50% cassava with 50% high energy concentrate diets were able to reach
market weight of 2 kilograms or more at 42 days of age. However birds fed the 70% sweet
potato with 30% low energy concentrate did not reach market weight for both the CLTC and
LDS sites. At the OTDF site birds were able to reach market weight when fed higher amounts
of sweet potato, possibly due to a more suitable environment for the birds.
(d) Dilution of a commercial diet with locally available food products
In some Pacific countries considerable quantities of copra meal are available for use in
poultry feeds. Pandi (2005) examined the growth rate of village broilers that were fed with a
commercial finisher feed diluted with 20-80 % of copra meal. In figure 2 diet 1 was 100%
broiler finisher (BF); diet 2, 80% BF + 20% copra meal (CM); diet 3, 60% BF + 40% CM;
diet 4, 40% BF + 60% CM; diet 5, 20% BF + 80% CM and diet 6, 100% CM. Diluting a
broiler finisher diet with 20-40% copra meal result in similar growth as the control diet and
inclusion of 60% copra meal resulted in acceptable growth (Figure 2). The extent of dilution
that is practiced depends on the availability and cost of copra meal.
2500
Weight (grams)
2000
100%BF
80%BF + 20%CM
1500
60%BF + 40%CM
1000
40%BF + 60%CM
20%BF + 80%CM
500
0
21
28
35
42
49
53
Age (days)
Figure 2.
Weekly average weights of meat birds fed various ratios of copra meal and
commercial broiler finisher over the period 21-53 days (from Pandi, 2005).
150
Aust. Poult. Sci. Symp. 2010...21
III. SUSTAINABILITY
(a) Cost of imported feed
The major issue restraining the development of the smallholder poultry sector in the Pacific is
perceived to be the lack of regional small scale feed manufacturing plants and the high cost of
imported feed as well as cheap imports. Despite this there are adequate supplies of fishmeal,
cassava, sweet potato, fresh coconut and maize, which could form the basis of the feed
industry throughout the Islands. Poultry production is an important smallholder industry.
However the growth and expansion of the smallholder industry is dependent on reducing the
reliance on imported complete feed or feed ingredients from overseas. Smallholder poultry
production has been hampered or abandoned in many parts of the Pacific due to increased
feed costs. Cheap diets based on locally available ingredients and agricultural by-products
will encourage new poultry farmers into the industry and encourage others back to poultry
production. Use of local feeds will provide an opportunity to develop a sustainable system
that will support smallholders who have not benefited from such production systems.
Knowledge and skills in poultry feed diet formulation based on locally available ingredients
and agricultural by-products have improved rapidly in many Asian countries. Smallholder
poultry farmers in several locations around the Pacific appear to have demonstrated that they
can make a profit from these farming systems using imported complete feeds supplemented
with local feed ingredients or in some cases, using only local formulated diets.
(b) Milling equipment
In the Pacific Islands most of the commercial diets used to feed poultry have been imported
from New Zealand, Australia or the United States. Where feed mills have been established in
the Pacific they have relied on the use of imported feed ingredients, but many commercial
mills have closed due to the escalating costs of imports. Currently only small amounts of
local feed resources are used in commercial rations by the larger feed mills in the Pacific due
to the high cost of transport, storage difficulties, reliability of supply and constraints in feed
formulation because of variable quality. However the establishment of small scale regional
feed manufacturing centres (producing 5-10 tonne/week) in areas where local feed supply is
plentiful may overcome some of these issues. The lack of suitable feed making equipment to
produce diets is the major constraint limiting the development of the small-scale feed
industry in the Pacific. However such equipment is becoming progressively more available.
In PNG and Tonga progress is being made in establishing small scale feed manufacturing
centres which use local feed ingredients to produce lower cost pig and poultry diets.
Relatively cheap (approx. A$8000-10000) small scale feed mixing equipment can be
purchased to mill diets.
(c) Cost benefits of small-scale feed mills
Smallholder broiler farmers in PNG purchase day-old broilers usually in lots of 52 and grow
them out for 6 weeks using commercial feed and then sell them as live birds in local markets.
The cheapest carbohydrate source to feed broilers in the highlands is sweet potato while in
the lowlands cassava is preferred. Comparisons were made of the costs of the following
feeding options; 1) lowlands mini-mill HEC (50%) supplemented with cassava (50%); 2)
lowlands commercial HEC (50%) supplemented with cassava (50%); 3) lowlands
commercial feed (50%) diluted with copra meal (50%) and 4) lowlands commercial feed. In
the highlands the costs of the same feeding options were determined except sweet potato was
used instead of cassava. The lowland broiler farmers are close to the commercial mills while
151
Aust. Poult. Sci. Symp. 2010...21
the highland farmers have extra costs associated with transport of commercial feed. The
feeding costs in both the highlands and lowlands show advantages when concentrates (HEC
and LEC) are fed with sweet potato or cassava compared to use of commercial finisher
broiler feed based on imported ingredients (Black 2009, unpublished) – Table 5.
Table 5.
PNG village feeding costs in Kina (A$0.43) per batch of 52 meat chickens
grown to 6 weeks of age. (1Kina=AUD$0.43)
PNG feeding system
Lowlands mini mill HEC + cassava
Lowlands commercial concentrate + cassava
Lowlands diluted commercial feed
Lowlands commercial feed
Highlands mini mill LEC + sweet potato
Highlands commercial concentrate + sweet potato
Highlands diluted commercial feed
Highlands commercial feed
Kina
276
318
334
431
311
350
359
481
The proportion of the costs of the concentrate and dilution feeding system relative to using
commercial feeds alone show that there is a strong cost reduction incentive at the village level
to adopt alternative feeding strategies based partly or wholly on local feed ingredients (Table
6) .
Table 6.
Alternative PNG village feeding systems costs as a proportion of the commercial
feeding costs
PNG Feeding system
Proportion
Lowlands mini mill HEC + cassava/lowlands commercial feed
0.64
Lowlands commercial concentrate + cassava/lowlands commercial feed
0.74
Lowlands diluted commercial feed/lowlands commercial feed
0.77
Highlands mini mill LEC + sweet potato/highlands commercial feed
0.65
Highlands commercial concentrate + sweet potato/highlands commercial feed
0.73
Highlands diluted commercial feed/highlands commercial feed
0.75
It may be possible to achieve feed cost savings of up to 35%, with the mini-mill concept
being the most advantageous. The mini-mill concentrate was based on copra, fishmeal and
coconut oil.
AKNOWLEDGEMENT
These studies were supported by the Australian Centre for International Agriculture
Research.
REFERENCES
ALFID (2002) Australian Livestock Feed Ingredient Database. SARDI, Roseworthy, South Australia.
Glatz PC (2006) Poultry feeding systems in PNG. Final report to Australian Centre for International
Agriculture Research.
Glatz PC (2008) Improving the profitability of village broiler production in PNG. Annual report to
Australian Centre for International Agriculture Research.
Grayson R, Campbell F (2004) Improved Kokorako keeping. Kastom Gaden Association and
TerraCircle Association Inc.
Pandi J (2005) PNG Journal of Agriculture, Forest and Fisheries, 48, No 1 & 2.
152
Aust. Poult. Sci. Symp. 2010...21
THE ROLE OF THE WORLD’S POULTRY SCIENCE ASSOCIATION (WPSA) IN
SUPPORT OF POULTRY PRODUCTION IN DEVELOPING COUNTRIES
R.A.E. PYM 1
Summary
Over the past thirty years WPSA has played an increasingly active role in support of poultry
production in the developing countries. The recent and projected dramatic increases in
poultry production and consumption in many of the developing countries argues for the
establishment of sound scientific and technical support to backstop and facilitate an efficient
and sustainable poultry industry in these countries. Poultry production from all sectors has the
potential to impact positively and meaningfully upon the United Nations’ Millennium
Development Goals of alleviation of extreme poverty and hunger in the developing world,
despite the mitigating effects of climate change, biofuel production and HPAI. Through its
involvement in the organisation globally of meetings on poultry science and technology and
on all manner of industry-related issues, the establishment of WPSA branches in developing
countries, its close cooperation with FAO, its support for the establishment of poultry
networks in developing countries, and its involvement in poultry development projects,
WPSA has played a meaningful role in support of poultry production in many of the
developing countries of the world.
I. INTRODUCTION
In light of the recent global economic downturn, it is difficult to predict future global
production and consumption of commodities generally and of poultry meat and eggs in
particular. Projections (Farrell, 2008; Speedy, 2003) based on FAO statistics from 2000 to
2006, suggest that chicken meat production will increase by 2.3 and 4.0% per year
respectively in developed and developing countries between 2006 and 2016, and egg
production will increase by 0.1 and 3.5% respectively. Much of this increase will be seen in
the more affluent, larger population developing countries. Similarly, estimates were made of
the increase in feed required for poultry meat and egg production in developed and
developing countries over the same period. The increase in feed required for poultry meat
production was 2.5 and 18.0% per year respectively and for egg production was 2.0 and
12.3% respectively. The potential impact of biofuels on the significantly greater increase in
feed requirements in developing countries over the next 8 years, gives considerable cause
for concern. Farrell (2008) estimated that some 250 million hectares of arable land will be
required globally by 2016 to annually grow the 1100 million tonnes of grain and protein
crops to meet the need for feedstuffs for compounded feeds for all livestock.
Although world human population growth has slowed down to about 1.1% per year,
there will still be another 700 million mouths to feed by 2016, the large majority of these in
developing countries. Africa’s population is expected to increase by about 2.4% from about
890 million at present, to 1050 million by 2016. Growth in the developed world will be
almost zero; estimates for the population increase in the EU to 2016 are only about 0.1%
(Anonymous, 2007).
1
36 Whitmore Street, Taringa Q. 4068
153
Aust. Poult. Sci. Symp. 2010...21
There is no doubt that affluence is the force driving livestock production; as
developing countries become more wealthy, the populace demands more animal products
and poultry production is often the first livestock industry to respond to that demand. The
technology is universal and portable, easily transferable, and eggs and chicken meat are
generally free from cultural taboos. Demand in China and India for livestock products has
been increasing at a rapid rate in concert with rapid economic growth (at least up until the
recent economic downturn). Other countries with large populations, such as Indonesia,
Vietnam and Brazil have recently experienced rapid economic growth and have a large
expanding middle class. It is thus not surprising that the future demand for animal products
will come from developing countries and comparatively little from developed countries.
However, many of the very poor countries such as those in sub-Saharan Africa, West Asia
and North Africa, even prior to the global economic woes, have experienced a decline in
energy and protein from animal products.
II. POULTRY PRODUCTION SYSTEMS IN DEVELOPING COUNTRIES
FAO has categorised the different poultry production systems which operate in most
developing countries into four sectors. These range from sector 1 large scale commercial
operations, often integrated with hatcheries, feed mills and processing plants, and utilising
sophisticated housing and equipment, with confinement rearing/housing and feeding of
commercial layer or broiler genotypes, through to sector 4 small-scale family semiscavenging flocks of indigenous genotype birds. The former sector is essentially identical to
production systems seen in most of the developed countries and the feed used is often
compounded from imported feedstuff ingredients. Sector 2 systems tend to be similar to
sector 1, but are generally smaller, not integrated to the same extent with other poultry
facilities, and make greater use of local materials in shed construction and equipment. They
also tend to employ manual rather than automated systems of feeding and egg collection.
For both sectors 1 and 2 broiler operations, chickens are usually processed centrally before
distribution and sale.
Sector 3 production units are essentially small-scale commercial units of from 100 to
500 commercial broilers or layers with confinement housing and feeding using
commercially compounded feed. Houses and equipment are often constructed from local
materials and eggs and live birds are usually sold at local markets, rather than through
distributors or directly into regional markets. Whilst sector 3 production can be profitable in
the short-term in niche markets, competition with the larger commercial operators for
chickens, feed, pharmaceuticals and markets, puts considerable strain on the long-term
viability of such operations.
By way of contrast with the above commercial systems, sector 4 is largely comprised
of very small flocks of 10 to 50 indigenous breed birds kept by a large proportion of families
in villages in the rural regions of many countries. These birds receive household scraps and
scavenge around the village during the day; as such feed costs are minimal. The hens hatch
their own eggs and rear the chicks to an age when they can fend for themselves. These small
poultry flocks make a meaningful contribution towards poverty alleviation and household
food security in rural communities (Gueye, 2000). They also account for a significant
proportion of poultry meat consumed in many of the developing countries (Pym et al. 2006)
Productivity in sector 4 production systems is low, but so too are inputs. The hen
usually lays only 40 to 60 eggs per year over 3 to 4 clutches, but spends much of her time
hatching the eggs and rearing the chicks. Probably the greatest constraint to productivity and
profitability of this system, is the usually very high attrition rate in young chicks (often more
than 60%) due to predation, disease, malnutrition and climatic stress. In most countries,
154
Aust. Poult. Sci. Symp. 2010...21
there tends to be a preference for chicken meat than for eggs, and consequently, most of the
eggs tend to get set under the hen to produce chicks. Whilst hatchability is normally quite
good, the high attrition rate means that only a few of these chicks survive to an age when
they can be eaten, or in the case of females, to sexual maturity.
There are opportunities for improving productivity and sustainability in all four
production systems, but it is important to understand the constraints under which each
system operates. For poor people living in rural regions of many countries, the only poultry
meat and eggs they are ever likely to eat are those they produce themselves from their small
family flocks of indigenous chickens. However, because of limited availability of a
scavenging feed resource base, and concerns about the transmission of diseases (particularly
HPAI) to people and to commercial poultry flocks, there are relatively few remaining sector
4 scavenging flocks in urban or peri-urban communities in most countries. There is no
question that the most efficient and effective means of providing poultry meat and eggs to
people in urban and peri-urban areas, is from commercial production from broilers or layers,
whether it be from sector 1, 2 or 3 production systems. Household income in these
communities is for the most part significantly higher than in rural communities and poultry
meat and eggs are moderately affordable by many.
III. THE ROLE OF WPSA AND ITS CONTRIBUTION TOWARDS POULTRY
PRODUCTION GLOBALLY.
The World’s Poultry Science Association (WPSA) has over seven thousand members in 75
countries around the world. The objectives of the association are to promote the advancement
of knowledge of all aspects of poultry science and the poultry industry world wide,
principally by facilitating exchange of information through the organisation of group
meetings, regional conferences and World Congresses. To promote membership of the
organisation in developing countries, the cost of belonging to the world Association in those
branches, is only half that of the membership of branches in the developed countries. All
members receive copies of the World’s Poultry Science Journal, published quarterly and now
in its sixty-first year of publication.
The main international forums organized by WPSA are the World’s Poultry
Congresses, held every four years. World Congresses include a wide ranging scientific and
technical program and the meeting also incorporates a large and comprehensive poultry
industry trade exhibition. The meetings are very well attended by a broad cross section of
people from the scientific community, poultry industry, educational institutions and
government. The last meeting (the 23rd Congress) was held in Brisbane, Australia in July
2008 and the 24th Congress will be held in Salvador, Brazil in 2012.
There are presently two Federations of the Association: viz the European and Asian
Pacific Federations. Each Federation stages a conference every four years, approximately
midway between the World Congresses. Additionally to this, the European Federation, has
some 11 working groups on Economics and Marketing; Nutrition; Breeding and Genetics;
Egg Quality; Poultry Meat Quality; Reproduction; Poultry Welfare and Management; Turkey
Production; Education and Information; Physiology; and Ratites, whilst the Asian Pacific
Federation, established somewhat later, has two working groups on: Small-scale Family
Poultry Farming; and Waterfowl. Each of the working groups has the responsibility for
organizing regular symposia in the relevant area. These symposia are recognized
internationally as the main global forums for scientific and technical discussion in the
respective areas and are for the most part very well attended by industry.
155
Aust. Poult. Sci. Symp. 2010...21
The nature and scope of the above meetings and the close involvement of the poultry
industry, means that they have had, and will continue to have, a significant impact on
improvements in the efficiency and sustainability of the global commercial poultry industry.
The extent to which this impacts upon the wellbeing of the poor in developing countries, is
influenced by many factors. An important area of involvement with potential to impact more
directly upon poverty and household food security, is the work in support of small-scale
family poultry farming.
IV. WPSA SUPPORT FOR SMALL-SCALE FAMILY POULTRY FARMING
The commitment to improving the efficiency and sustainability of small-scale family poultry
farming was very much in evidence in the scientific and technical programme at the 19th
World’s Poultry Congress in Amsterdam in 1992; this commitment has continued in
subsequent Congresses.
A meeting was held in the family poultry farming session at the 7th Asian Pacific
Federation Conference at the Gold Coast, Australia in October 2002 where a proposal to
establish a Working Group on Family Poultry Farming was approved by the assembled
membership of WPSA. The inaugural meeting of the Working group was subsequently held
in March 2005 in conjunction with the 4th International Poultry Show and Seminar in Dhaka,
Bangladesh. At that meeting WPSA funded the attendance of two delegates each from Sri
Lanka and Indonesia to provide accounts of the impact of the 2004 Boxing Day tsunami on
the poultry industry in the respective regions, with the aim of developing proposals for
assistance.
In keeping with the commitment to support small-scale family poultry development,
at the 11th European Poultry Conference in Bremen in October 2002, the WPSA Board
endorsed the proposal to include the International Network on Family Poultry Development
(INFPD) as a global working group within WPSA. Discussions regarding the most effective
and efficient structural arrangements for this cooperative arrangement are on-going, and will
be a focus of the WPSA presidency of the writer.
In March 2007, the WPSA Asian Pacific Federation working group on Small-scale
Family Poultry Farming convened a symposium on The impact of Avian Influenza on Smallscale Family Poultry Farming in Developing Countries in conjunction with the 8th Asia
Pacific Poultry Conference in Bangkok, to which a large number of internationally
recognized family poultry researchers and promoters were invited. The meeting was
convened with the express purpose of providing objective information about the impact of
HPAI on small-scale family poultry farming (SSFPF) production and the role of SSFPF in
transmission of the disease, following the demonization of SSFPF by a number of
governments. The other important aim was to identify appropriate measures that need to be
taken by SSFP keepers to minimize the effects and transmission of the disease. Attendance of
the 30 or so speakers was jointly sponsored by FAO, CTA and WPSA. Additional to papers
dealing with a wide range of technical and procedural issues, there were regional reports on
the then current status of HPAI and small-scale poultry farming in Lao PDR, Thailand,
Bhutan, Burma, Papua New Guinea, The Philippines, Malaysia, Cambodia, Indonesia, India,
Sri Lanka, Bangladesh, Vietnam, Mozambique and Africa (generally).
At the 23rd World’s Poultry Congress held in Brisbane, Australia in July 2008, one of
the eight main concurrent streams was on Poultry Production in Developing Countries. There
were some 40 papers presented in the stream and the final session within the stream involved
a workshop aimed at facilitating cooperation and collaboration between all stakeholders
involved in support of small-scale family poultry production. The report from the workshop
and the 15 invited papers in the stream was published in the June 2009 edition of the World’s
156
Aust. Poult. Sci. Symp. 2010...21
Poultry Science Journal. A recommendation from the workshop report was that an overarching entity “Poulet Sans Frontiers”, mooted by the writer in an AusAID-sponsored
conference in Tanzania in 2005 and with membership from all agencies and organizations
supporting family poultry farming, be established. The collective desire is that this body will
impact meaningfully on the level and nature of global support for small-scale family poultry
farming.
V. POULTRY PRODUCTION IN DEVELOPING COUNTRIES
AND THE MILLENNIUM DEVELOPMENT GOALS
The relatively recent problems of climate change, biofuel production (Lyons, 2007) and the
global economic uncertainties with their attendant effects upon global agricultural
production, have made the United Nations’ Millennium Development Goals, announced in
2000 with a range of targets set for 2015, considerably more difficult to attain. The
Millennium Development Goals are broad and multi-faceted and address all elements of
desirable development. Although not fully comprehensive, improvement in the efficiency and
sustainability of poultry production in developing countries has the potential to have a
significant impact upon a number of the Goals, particularly the primary goal of eradication of
extreme poverty and hunger. An additional mitigating factor here relating specifically to
poultry production, is the effect of HPAI.
The mechanisms by which improvements in poultry production efficiency can impact
upon poverty alleviation, are quite different for urban dwellers, who are largely dependent
upon purchased poultry meat and eggs, and rural villagers who mostly produce their own
meat and eggs from their small scavenging flocks of indigenous birds. In the former case, the
price of meat and eggs is very much in the hands of private enterprise and is influenced by
many external factors including government regulation. Since feed accounts for about 70% of
production costs, the impact of feed prices, as influenced by global markets, climate change,
biofuel competition etc, on the price of meat and eggs from commercial broilers and layers, is
profound. High feed prices present real challenges to the governments in question and to
donor agencies if the negative impact of this on the attainment of the MDGs is to be averted.
The absence of the need to provide compounded feed to small-scale rural scavenging
poultry flocks, means that they are significantly more buffered to the effects of the factors
influencing feed prices. The critical limitation here, however, is the generally low level of
productivity of the indigenous birds, the risk-averse attitude of the farmers to adoption of
improved technologies, and as a consequence, the commonly high levels of attrition in young
birds and mortality in the older birds. It has been demonstrated (Pym, 2005) that appropriate
and simple cost-effective interventions can have a significantly positive impact on the
profitability of small-scale family poultry production resulting in the alleviation of poverty,
improved household food security and a reduction in malnutrition, improved educational
opportunities for the children, and empowerment of women as the principal poultry keepers
in most communities. Small-scale poultry production offers the opportunity for the first step
out of poverty; to allow choices to be made as to further avenues of income generation
(Alders and Pym, 2008). The aim is certainly not to turn all rural families into small or
medium-scale commercial poultry producers; this would be counter-productive. A small
percentage may choose to go down this track.
There is thus the opportunity for improvement in the efficiency and sustainability of
poultry production to have a beneficial impact upon a number of the Millennium
development Goals including: the eradication of extreme poverty and hunger; education of
the children; malnutrition and hence childhood mortality; and the empowerment of women.
In households where there is a lack of able-bodied workers, such as households affected by
157
Aust. Poult. Sci. Symp. 2010...21
HIV/AIDS or those that have a disabled family member, village poultry provide a source of
high quality nutrition and income without requiring much in the way of labour or financial
inputs (Alders and Pym, 2008). Poultry keeping is an easy activity that can contribute to
household food security and income. Given that women are the main carers of sick people
and that the poultry are usually under women’s control, poultry can play an important role
providing them with additional resources to carry out their important task of supporting
people living with HIV/AIDS (IRPC, 2005).
VI. ACROSS-SECTOR SUPPORT FROM WPSA FOR POULTRY
PRODUCTION IN DEVELOPING COUNTRIES
Following on from the FAO sponsored “Poultry in the 21st Century” meeting in Bangkok in
November 2007, it became apparent that there was a need for a broad thrust to facilitate
efficient and sustainable poultry production from all sectors of the industry in developing
countries, not just small-scale sectors 3 and 4. WPSA has worked in close cooperation with
FAO in this endeavour. In the preparation of the FAO’s State of Food and Agriculture 2009
report, the writer was asked in 2008 to coordinate a review on Technological Change and its
Impact on Poultry Development, covering genetics and breeding, disease prevention, housing
and environmental control, processing of poultry meat and eggs, with input from five
Australian and overseas WPSA members.
More recently, the writer was asked by FAO to assist them by coordinating a series of
reviews to include in their revised AGA “Poultry Production in Developing Countries”
website. The focus of the site is to be broadened from just small-scale poultry production, to
all aspects of poultry production across all sectors of the industry in developing countries.
Most of the contributors to the twelve component reviews, are members of WPSA. The
reviews are at the time of writing, in an advanced stage of preparation and the website should
be operational by the time of the 2010 Australian Poultry Science Symposium.
(a) WPSA Mediterranean and African poultry networks
In recognition of the need to establish structures to facilitate the advancement of poultry
science and the development of the poultry industry in developing countries, WPSA has been
involved in two notable developments over the past twelve months. In June 2009 the Board
of WPSA endorsed the establishment of the Mediterranean Poultry Network which will focus
on the promotion of poultry science and the development of the poultry industry throughout
the Mediterranean region in Southern Europe, the Middle East and Northern Africa. The
Network will operate initially under Working Group 11 (Education and Information) of the
WPSA European Federation. One of the particular aims of the Network is to provide support
to the poultry industry in the countries of Northern Africa and the Middle East, that are not
members of the European Federation or of its working groups. The main forum for scientific
and technical discussion between members of the Network will be at the biennial
Mediterranean Poultry Summit meetings, already established. The most recent Summit was
held in Antalya, Turkey in October 2009 and the next meeting will be held in Egypt early in
2012.
There have been ongoing discussions for a number of years about the possibility of
forming an African Federation of WPSA. It has proven to be a difficult task to bring this to
fruition, largely because of the size of the continent, difficulties in identifying suitable
working group areas/titles, and the difficulty and cost of bringing people together at one
location. To address some of these problems, a proposal was made recently for a virtual
(web-based) African Federation of WPSA. Further discussion by the Board of WPSA has
resulted in the Board’s endorsement of the development of a sub-Saharan African Poultry
158
Aust. Poult. Sci. Symp. 2010...21
Network whose focus would be on the promotion of poultry science and the development of
the poultry industry in the region, and whose main mode of communication and interaction
between members, would be via the Internet. To initiate the formation of the Network and
bring the key players together, a workshop is to be held in France, immediately following the
European Poultry Convention in Tours in late August 2010. The plan is that delegates to the
workshop will attend EPC prior to their meeting. A steering committee has been established
and the program for the workshop is in preparation. FAO have provided considerable
encouragement, both verbal and material, for the conduct of the workshop and the
establishment of the Network.
It is important to note that the proposal is NOT focused only on small-scale village
poultry farming; the focus is on giving the African members of WPSA (scientists and
industry personnel) an opportunity to meet at a first-class, broad-based poultry science and
technology meeting in Europe to (apart from attending this) spend some time discussing the
constraints to the efficiency of commercial and "village" poultry meat and egg production in
Africa, and how poultry can play a bigger role in meeting the protein needs of the burgeoning
population in the developing countries of Africa.
(b) Poultry Think Tank Meeting.
A “think-tank” meeting was held in Freising, Germany on 19th June 2009 to discuss social
equity and sustainability issues relating to the present systems employed globally in the
production of poultry meat and eggs. Participation at the meeting was by invitation and
included representatives from the poultry industry, the poultry research community, FAO and
WPSA. Representatives from WPSA were Roel Mulder (Secretary), Piet Simons (Past
Secretary) and the author.
The meeting arose from an invitation issued by Dr John Hodges, an agricultural
researcher with interests in ethical issues related to animal production, during the presentation
of his keynote address “Emerging boundaries for poultry production: Challenges,
opportunities and dangers” at the 23rd World’s Poultry Congress in Brisbane on 30 June
2008. He challenged the industry to examine its practices from social equity and
sustainability perspectives. Dr Dietmar Flock, Poultry breeder and Past President of the
European Federation of WPSA, subsequently initiated discussions with Dr Hodges which led
to the organization of the think tank meeting in Freising. A full report on the outcomes of the
meeting will shortly be published in WPSJ. The draft report is presently with the meeting
participants for their comments and suggestions.
One of the recommendations from the Think Tank was that better use should be made
of the technical expertise in WPSA, FAO and industry as advisers to development projects
involving poultry in developing countries. Poultry often form part of agricultural
development projects, but not infrequently poultry production expertise in the NGO team is
quite limited, and poor advice is not uncommon. It should be noted that this is not restricted
to the smaller NGOs. WPSA needs to be proactive in establishing links with development
project funders globally.
159
Aust. Poult. Sci. Symp. 2010...21
REFERENCES
Alders RG, Pym RAE (2008) Village Poultry: Still important to millions eight thousand years
after domestication. Proceedings 23rd World’s Poultry Congress, Brisbane, Australia.
30 June-4 July 2008. WPSA CD ROM
Anonymous (2007) European Commission for Directorate General for Agriculture and Rural
Development July 2007. Agricultural Market and Income in the European Union
2007-2014
Farrell DJ (2008) The future eaters. Proceedings 23rd World’s Poultry Congress, Brisbane
Australia. 30 June-4 July 2008. WPSA CD ROM.
Guèye EF (2000) The role of family poultry in poverty alleviation, food security and the
promotion of gender equality in rural Africa. Outlook on Agriculture 29,129-136.
IRPC (2005) First Interim Activity Report to FAO. Improvement of village chicken
production by junior farmers and people living with HIV/AIDS. Component of
OSRO/RAF/403/SAF. Manica and Sofala Provinces, Mozambique. International
Rural Poultry Centre, Brisbane, Australia.
Lyons P (2007) World Poultry 23, 20-22.
Pym RAE (2005) Research on small-scale family poultry improvement: An overview.
Proceedings 4th International Poultry Show and Seminar. Dhaka, Bangladesh. March
2005. Pp. 147-153.
Pym RAE, Guerne Bleich E, Hoffmann I (2006) The relative contribution of indigenous
chicken breeds to poultry meat and egg production and consumption in the developing
countries of Africa and Asia. Proceedings 12th European Poultry Conference, Verona,
Italy. 10-14 September 2006. WPSA CD ROM.
Speedy AW (2003) Journal of Nutrition 133, 4048S-4053S.
160
Aust. Poult. Sci. Symp. 2010...21
THE EVOLVING ROLE OF FAO IN POULTRY DEVELOPMENT
S.D MACK 1
Summary
Both the global poultry sector and FAO’s involvement have changed significantly over the
last 50 years. Enormous advances in poultry productivity have occurred and with modern
husbandry chicken has become the most effective terrestrial animal in converting feed grain
into meat. Yet the poultry sector still remains highly diverse with a growing dichotomy
between the industrial, commercial production and extensive back-yard poultry systems.
Small-scale poultry keeping continues to make a significant contribution to the livelihoods of
some of the world’s poorest communities and, as such, remains an important tool in rural
poverty reduction programmes. The growth of intensive poultry production has given rise to a
range of issues concerning food safety, environmental pollution, biodiversity and animal
welfare. FAO meanwhile has moved away from direct technical assistance and traditional
development projects, with the exception of its emergency and disaster response programme,
to a more supporting and informing role. The current focus is on the interaction and impact of
poultry production with the international public goods such as poverty reduction, public
health and natural resource management. This paper reviews the link between poultry
production and these international public goods. It then provides an overview of the range of
activities that FAO undertakes to support the global poultry sector and draws heavily on
material represented at an FAO sponsored conference in Bangkok in November 2007
(http://www.fao.org/AG/againfo/home/events/bangkok2007/en/index.html).
I. INTRODUCTION
Within the lifetime of FAO, the global poultry sector has undergone unprecedented change.
Poultry meat has moved from a luxury food item to an abundant, affordable and accessible
source of quality protein in the diet. Whilst the most visible structural changes have been seen
in the developed economies, the consequences on how poultry products are produced,
processed, marketed and consumed are evident across the globe. The poultry sector has
evolved into the fastest growing, most efficient and indeed the most flexible of all the
livestock sectors (Upton, 2008).
The rapid growth in the poultry sector has been driven by both supply and demand
factors. Demand has been driven by population growth; increases in per capita incomes;
urbanization, and changing consumer preferences (Narrod et al., 2008). There remains a
positive relationship between income and the consumption of food with a high income
elasticity, such as poultry products, and this has traditionally involved a shift from starches to
meat.
On the supply side, major technological advances have resulted in a dramatic shift
from labour intensive to capital intensive systems, resulting in specialization, greater
biological efficiency, economies of scale and vertical integration within the supply chain.
Advances in animal breeding have resulted in major increases in production efficiency. In
commercial broiler production feed conversion ratios (FCR) have reached 1.6:1.0 making the
modern hybrid broiler the most efficient terrestrial farm animal to utilize feed grain. Likewise,
the modern hybrid layer is capable of producing 330 eggs per year. Genetic improvement has
1
Senior Animal Production Officer, Animal Production and Health Division, FAO, Rome
161
Aust. Poult. Sci. Symp. 2010...21
gone hand-in-hand with a greater understanding of poultry nutrition and the pathology of
poultry diseases. Modern husbandry practices (feeding systems, animal health care, housing
and control of the micro-environment) have been developed that utilize these and resulted in a
strong, integrated and international poultry industry.
The poultry sector is however far more diverse than just the large-scale commercial
operations that provide affordable poultry products to the growing urban communities. In the
rural areas of developing countries, the ubiquitous free-ranging chicken represents a
widespread, extensive, ‘low-input, low output’ production system which provide
supplementary income and food to the household, an asset and a means to meet social
obligations (Dolberg, 2008). There are also many small to medium-scale poultry producers
have also exploited the wide availability of commercial hybrids and compound feeds to
develop successful enterprises. There is a concern over the longer-term future of these small
to medium-scale enterprises. Unlike the backyard systems, they are dependent on purchased
inputs but do not have the economies of scale of the industrial poultry sector. Furthermore it
is expected that their transaction costs will increase as they have to abide with more and more
public health and trading standards.
II. POULTRY AND INTERNATIONAL PUBLIC GOODS
The FAO Animal Production and Health Division (AGA) anchors it programme on three,
interrelated, global international public goods and how they relate to livestock. These include
(i) social equity (poverty alleviation), (ii) veterinary public health (animal health in its broader
context) and (iii) sustainable natural resources (animals and land, water, air, biodiversity,
ecology).
a) Poultry and social equity
There is a growing recognition amongst the development community of the contribution of
small-scale poultry production in reducing poverty, enhancing household food security and in
promoting of gender equality (McLeod, 2008). Traditional back-yard poultry production
systems remain essentially ‘low-input, low output’ systems that utilize family labour (women
and children) and characterised by rudimentary supply chains. Products are either consumed
in the household or sold locally and poultry are also an important asset to meet social
obligations. With extremely high losses of chicks before they reach a productive age from
disease (notably Newcastle Disease) and predation, there is ample scope for increasing
productivity through loss reduction. Simple housing to protect chicks from predators and the
elements, supplementary feeding, and vaccination against Newcastle Disease can have a
substantial impact. Reducing losses also reduces the requirement for hatching eggs to replace
the lost birds which immediately release eggs to be sold or consumed (Sonaiya, 2008). It has
to be said that, even with the advent of heat stable Newcastle Disease vaccines and numerous
project interventions, relatively little impact has been made in increasing the productivity of
these systems. These systems while important in supplementing and maintaining livelihoods,
are unlikely in themselves to provide a viable mechanism lift households out of poverty.
There are a number of notable exceptions. The FAO Village Poultry Groups in
Afghanistan (see III i) and the Bangladesh Model. The Bangladesh Model consists of an
integrated system of production, marketing, input-supply and service provision supporting a
network of individual, specialised producers and entrepreneurs (Dolberg, 2008). First initiated
in 1983 it has attracted substantial donor support and driven by the Department of Livestock
Services and BRAC (Bangladesh Rural Advancement Committee). Evaluation studies have
demonstrated that the approach has a strong ‘pro poor’ bias and had a significant impact of
the economic and nutritional status of the poor – especially women and children. Despite
162
Aust. Poult. Sci. Symp. 2010...21
concerns over the sustainability of the programme after the withdrawal of donor support;
there is little doubt that the programme has demonstrated the potential of small-scale poultry
as a tool for improving rural livelihoods.
Semi-commercial production systems depend far more on a functioning, if informal,
supply chain for reliable access to inputs (day-old-chicks, feed, drugs etc.) to exploit local
markets and market preferences where traditional products may command a premium.
Growing rural populations, improve infrastructure (roads, transport and communications) are
opening up new market opportunities. There are many examples, especially in Asia, of
successful small to medium-scale poultry entrepreneurs. While the back-yard and the largescale commercial operations are largely segmented and are likely to remain so, there is an
increasing overlap and competition between the modern large-scale commercial sector and the
medium to small-scale producers. This happened in Thailand where, over two decades, the
share of small-scale production has fallen from 95 to 10 percent (Vinod and Arindam, 2008).
b) Poultry and public health
The poultry sector is associated with a range of public health risks arising from zoonotic
diseases notably, but not exclusively, Highly Pathogenic Avian Influenza (HPAI) and product
contamination. The poultry industry is dependent on maintaining the trust of the consumer
that it provides a safe and wholesome product. The dramatic, albeit short-lived, crash of the
poultry market in the aftermath of HPAI due to the loss of consumer confidence illustrates the
point.
Risks associated with food poisoning, notably from Salmonella and Campylobacter
spps. have also focused public attention on the safety of poultry meat and eggs. In addition to
the microbiological risks, chemical contamination of the feed (dioxins, heavy metals,
pesticide residues) can also enter into the food chain and the misuse of antibiotics in the feed
can result in increasing drug resistance. International guidelines and standards, such as Codex
Alimentarius, exist in addition to a plethora of public and private sector standards. In
developed countries, such food safety regulations are largely enforced and the poultry
industry is well motivated to minimize risk and HACCP procedures are well established and
mainstreamed. This is not always the case in developing countries were regulations, if
present, are extremely difficult to enforce. Consequently, food-borne illness remains a serious
public health issue in these countries and a major concern to international agencies such as the
World Health Organization (WHO) and FAO.
c) Poultry and sustainable resource management
The growth of the commercial poultry sector has resulted in geographical concentration and
intensification of production in ‘landless’ systems. Intensive production favors those areas
with access to cheap inputs and services (feed processors, slaughter houses, transport and
utilities) - such conditions are found in the vicinity of urban conurbations. This separation of
poultry production from its agricultural base and its move to areas requiring little or no
agricultural land can have a high environmental impact (Gerber et al., 2008).
At the production level there are local issues with smell, flies and a waste disposal.
Birds raised in intensive conditions consume high protein and high energy diets which give
rise to a manure high in nutrients such as nitrogen, phosphorus, and other excreted substances
(heavy metals, drug residues) introduced through the feed. Generally poultry manure is
regarded as a valuable commodity and is recycled as an organic fertilizer or used in animal
feed. It can however lead to local cases of nutrient based pollution which can result in
eutrophication of surface water, pollution of the ground water (nitrate leaching) and the
emission of nitrous oxide. Some metals (arsenic, cobalt, copper, iron, manganese, selenium
and zinc) that may be used as growth promoters or as prophylactic feed additives, and manure
163
Aust. Poult. Sci. Symp. 2010...21
can be a significant source of environmental contamination of these metals. There are also
issues with drugs and hormones entering into various food chains (Kiilholma, 2008).
The poultry sector is a major consumer of cereals (estimated at eight percent) and
soyabean meal (Hinrichs and Steinfeld, 2008; FAOSTAT). There is therefore an indirect
global dimension to the environmental impact of poultry production. The environmental
impact of feed production include those associated with intensive agriculture (misuse of
fertilizer, pesticides and herbicides), inappropriate crop expansion (deforestation),
biodiversity (including poultry breeds), associated greenhouse gas (carbon dioxide and nitrous
oxide) emissions and overexploitation of natural resources, notably fish stocks. The carbon
footprint of intensive poultry also has to take into account the relatively high level of energy
use associated with production, slaughtering, processing as well as those associated with
national and international trade of poultry products and inputs.
Concern has also been expressed regarding the erosion of biodiversity within the
poultry sector. Highly selected hybrid broilers and layers, controlled by a handful of breeding
companies, now dominate commercial production. Indigenous breeds represent the majority
of the poultry genetic biodiversity, they are well adapted to local conditions, especially those
found in some of the most resources poor communities in the world; yet many are under the
threat of extinction.
III. FAO SUPPORT TO POULTRY DEVELOPMENT
FAO addresses these public good issues through a) its field programme which provides direct
development support to its member countries, and b) its normative programme which
examines and informs on the broader issues associated with livestock and poultry production.
a) Field Programme
FAO has, since its inception, implemented projects that directly assist farmers and institutions
in developing countries. Projects can be classified according to their funding source. Trust
Fund projects are those implemented with external donor funding. The Technical Cooperation
Programme (TCP) funds projects from FAO’s core funds and aim to plug a specific technical
gap; they are catalytic in nature and with a short duration (maximum two years). Telefood is a
programme funded by public support (annual charity events) and provides small grants to
farmer groups to assist them establish or expand an enterprise – poultry keeping is a
population choice.
The nature of such assistance continues to change. Up until 15 years ago, every
technical division in FAO managed a large portfolio of development projects. The situation
began to change when the United Nations Development Programme (UNDP) strategy
changed to have its projects managed nationally rather than by the UN specialized agencies
such as FAO. This lead to a decline in the FAO field programme, since UNDP was a major
source of funding, and it is now dominated by TCP and other donor funded projects. There
has also been a major shift away from longer-term development assistance to shorter-term,
input focused emergency assistance programmes. Of the current portfolio of 481 projects
(TCP, Trust Fund and Telefood), where AGA is the Lead Technical Unit, 166 (35 percent or
60 percent excluding Telefood projects) are classified as emergency projects.
Poultry are a component in many FAO projects; however it is difficult to disaggregate
the information since poultry activities are often subsumed within broader development or
emergency assistance projects. The electronic database of FAO projects currently goes back
to 1980. Table 1 indicates the number and type of projects with a) poultry in the title and b)
livestock in the title. This is purely indicative and underestimates the actual number of
projects in which poultry activities can be found.
164
Aust. Poult. Sci. Symp. 2010...21
Table 1
FAO projects with ‘poultry’, ‘livestock’ or ‘avian’ in the title by funding
source 1980-2009
‘Poultry ‘in the title
‘Livestock’ in the title
‘Avian’ in the title
TCP
35
109
20
Trust Fund
30
131
109
Telefood
150
31
-
UNDP
7
35
4
Source: FAO internal data
Poultry figure strongly in the portfolio of emergency projects. Projects associated with
avian influenza (HPAI) feature significantly in the project portfolio since the late 2000’s.
However, many non avian influenza emergency response programmes also contain a poultry
component, usually in the form of restocking households who have lost stock, providing
supplementary feed or in some cases the provision of shelter. Poultry activities are also found
in some of the longer-term rehabilitation programmes in countries such as Afghanistan, Iraq,
Sudan, Gaza and the West Bank.
Indeed, one of FAO’s most successful poultry programmes has been in Afghanistan.
FAO started work in developing Village Poultry Producers Groups in 1999 in Faizabad
working with local women’s groups. The approach identified potential groups of women and
gave them practical training over a six month period including some limited inputs (chicks
and a small amount of mixed feed). The group leaders took on a service provider role giving
vaccinations, basic health care, selling feed and facilitating the marketing of eggs. The FAO
project continued to provide support to the group leaders for a year after the end of the initial
training period. Apart from the economic benefits and improvement in the household
nutrition, the programme had an important social impact in empowering women in their
communities. The programme has been expanded and today FAO is still supporting village
poultry production throughout Afghanistan with World Bank and IFAD support.
FAO support to its member countries is delivered primarily through its network of 136
country offices, five regional, 10 sub-regional offices and four FAO/OIE Animal Health
Officers. Each sub-regional office has a multi-disciplinary technical team which, in most
cases, includes a livestock specialist who is in turn supported by a livestock officer in the
regional office. At the Rome headquarters, AGA has around 30 professional officers at any
one time.
b) Normative Programme
All normative activities related to poultry provide directly and indirectly some degree of
policy support. AGA aims to inform decision makers (at all levels) on the important issues
they face, the options available to them and, importantly, their implications. In addition, it is
recognized that an enabling policy environment is essential for the development of a robust
and equitable poultry sector that provides producers with a fair return, the consumers with a
safe and affordable product, the private sector with a fair return on their investment, and the
public with a clean and safe environment. AGA, with support from DFID, has established a
Pro Poor Livestock Policy Initiative (PPLPI) with the aim to support the formulation and
implementation of livestock-related policies and institutional changes that have a positive
effect in the world’s poor. (http://www.fao.org/ag/againfo/programmes/en/pplpi/home.html)
Since keeping poultry is an important livelihood strategy for the world’s rural poor, the PPLPI
is of direct relevance. The PPLI approach compiles information and conducts analysis to
enhance pro-poor decision making within the livestock development community. Since
poultry keeping is an important livelihood strategy for the rural poor the PPLPI is of direct
relevance.
165
Aust. Poult. Sci. Symp. 2010...21
Given the importance and attention attached to livestock-related environmental issues
it is not surprising that AGA is working in this area. Notable was the publication of the
Livestock’s Long Shadow - Environmental Issues and Options (FAO, 2007a) which
successfully raised awareness to some of the negative environmental impacts caused by
livestock, and to gather the political will to address the issues. AGA also hosts the Livestock
Environment and Development Initiative (LEAD) (http://www.fao.org/agriculture/lead/en/)
which focuses on livestock issues related to deforestation, soil and water pollution, land
degradation and desertification. LEAD offers a number of decision support tools including the
Livestock and the Environment Toolbox that assists in assessing livestock environmental
interactions and models to predict nutrient balance and the direct and indirect consumption of
fossil fuels in various livestock production systems.
The erosion of genetic diversity in domesticated farm animals, including poultry,
especially in developing countries, is a serious concern to FAO. Its programme on managing
of animal genetic resources addresses policy and institutional concerns as well as technical
issues (http://www.fao.org/ag/againfo/programmes/en/A5.html). The State of the World’s
Animal Genetic Resources for Food and Agriculture (FAO, 2007b) identified significant gaps
in capacity to manage animal genetic resources. The international community responded in
2007 when it endorsed the Global Plan of Action for Animal Genetic Resources which
contains 23 strategic priority areas covering characterization and monitoring; sustainable use
and development; conservation; and policies, institutions and capacity building. AGA hosts
the Domestic Animal Diversity Information Systems (DAD-IS) as a definitive source of
information on the world’s species and breeds of farm animals (http://dad.fao.org/).
Poultry are commonly included in the international response to emergencies and
disasters. Poultry are seen as a quick and acceptable means of rebuilding the assets and
livelihoods of affected communities. However well meaning, not all interventions are
appropriate or have their desired impact. Hybrid birds distributed to village locations where
the management and inputs are inadequate inevitably do not survive. Distributing free feed,
drugs etc can also seriously disrupt the livelihoods of the local service providers. To address
the concerns, AGA working together with colleagues from the African Union, the
International Committee of the Red Cross (ICRC), the Feinstein Center (Tufts University) and
Veterinaries sans Frontier (VSF) developed and launched the Livestock Emergency
Guidelines and Standards (LEGS) (http://www.livestock-emergency.net/). LEGS takes a
livelihoods approach and provides basic standards, guidance, check lists and electronic
decision support tools for each of the major interventions: animal health, destocking,
supplementary feeding and water, shelter and restocking.
Information and knowledge are at the core of FAO’s work. The majority of AGA’s
publications and information are downloadable free, along with specialized pages, from its
website (http://www.fao.org/ag/againfo/home/en/index.htm). Over the last two years, AGA
has completed country profiles (http://www.fao.org/avianflu/en/poultryproduction.html) for
poultry in 27 countries which provide overviews of the main poultry systems, policies
statistics, reports and contact points. In early 2010, a new series of discussion and issues
papers on poultry development will be released on its poultry web page with the aim of
moving to a more interactive and participatory web portal later in 2010. Animal welfare is an
issue of interest to the poultry industry and in 2009 AGA launched the Gateway to Animal
Welfare gateway which is supported by 13 collaborating organizations (agencies and NGOs)
(http://www.fao.org/ag/againfo/programmes/animal-welfare/en/). The Gridded Livestock of
the World provides global maps for the distribution of the main species
(http://www.fao.org/ag/againfo/resources/en/glw/home.html) of livestock including poultry.
FAOSTAT (http://faostat.fao.org/) provides time-series data on trade, production and
consumption of poultry production and products for some 200 countries.
166
Aust. Poult. Sci. Symp. 2010...21
As an intergovernmental agency, FAO has close relations with the agricultural
departments in its 192 member countries. AGA also works closely with the private sector,
including the International Poultry Council (IPC), the International Federation of Feed
Industries (IFIF) and the International Federation for Animal Health; scientific organizations,
including the World Poultry Science Association (WPSA), as well as NGOs and sister UN
organizations. AGA was instrumental in establishing the International Network for Family
Poultry Development (INFPD) (http://www.fao.org/ag/againfo/themes/en/infpd/home.html)
which is an information exchange network to promote sustainable productivity within the
family poultry sub-sector.
Capacity building and training has always been central to FAO’s programme. With
financial support from IFAD, AGA is assisting the INFPD embark on a three year programme
to train young poultry professionals to give them experience and exposure to international
work. Each candidate will spend four to six weeks assigned as professionals in the FAO and
IFAD headquarters in Rome before they join an on-going project for six to nine months
practical experience where they will also carry out a specific assignment.
c) Animal Health
Protecting livestock against diseases and preventing disease spread is fundamental to AGA’s
programme in support of poverty reduction and veterinary public health. AGA’s animal
health programme, more than others, blurs the distinction between field and normative
activities. Responding to major epizootic diseases requires immediate direct assistance but
needs to be backed up by longer strategic planning, early warning and information systems,
and capacity building.
Of international concern is the transboundary nature of many animal diseases,
including zoonotics, which can quickly spread from the farm, through markets, to the entire
country and beyond. Highly Pathogenic Avian Influenza is a notable, but not the only,
example. AGA is responsible for the animal health component of the Emergency Prevention
System for Transboundary Animal and Plant Diseases (EMPRES) which was established by
FAO in 1994 http://www.fao.org/ag/againfo/programmes/en/empres.html, EMPRES provides
information, guidance and training to prevent, contain and control major disease outbreaks. It
is also responsible for the surveillance of newly emerging pathogens, which history has
shown often to originate from the poultry sector. The EMPRES approach incorporates early
warning, early detection, early reaction, enabling research, co-ordination and communication.
A Crisis Management Centre (CMC) has been established to ensure an effective and prompt
response to new disease outbreaks. It provides immediate technical and operational assistance
to governments facing a major disease outbreak
http://www.fao.org/emergencies/home0/emergency-relief-and-rehabilitation/cmc/en/ .
Responding to Avian Influenza (H5N1 HPAI) has been the greatest challenge facing
EMPRES. It was also a major economic shock and a wake-up call for the poultry industry and
attracted considerable public and media attention. Outbreaks have been recorded in 62
countries (five countries having reported only a single outbreak) in Europe, Asia, Africa and
the Near East, with the main foci being in Egypt, Indonesia and Vietnam. To date there have
been 258 human deaths from 442 reported cases (FAO AIDE News Update 62).
167
Aust. Poult. Sci. Symp. 2010...21
FAO, through EMPRES and with major donor support and in close collaboration with
OIE, has worked with all the affected and at-risk countries to strengthen national disease
intelligence, emergency preparedness, field surveillance and laboratory capacities, and has
developed and supported public awareness campaigns. Major studies and analysis have been,
and continue to be, undertaken on the epidemiology, economic and social consequences of the
disease and its control. Increasing biosecurity across all production systems and value chains
is an important focus within the programme primarily to control the spread of HPAI but
equally important for other important infectious poultry diseases (Sims, 2008). Numerous
national and international meetings of experts and stakeholders have been initiated or
facilitated by FAO since the first outbreak, which have resulted in important
recommendations and guidelines. In October 2008, as a result of a consultative process, FAO
and OIE in collaboration with WHO published The Global Strategy for the Prevention and
Control of H5N1 Highly Pathogenic Avian Influenza (FAO, 2008).
Early warning and the ability to predict the spread of a disease is essential for effective
containment and control of disease. The Global Early Warning and Response System for
Major Animal Diseases including zoonoses (GLEWS) (http://www.glews.net/) is a
collaborative effort between FAO, OIE and WHO to combine their early warning and
response mechanisms into an efficient tool.
IV. CONCLUSIONS
Growth in the scale, efficiency, market access and production of commercial poultry
production over the last 50 years is nothing short of phenomenal. As a result, poultry products
are now available and affordable. Yet, despite these advances, the poultry sector as a whole is
even more diverse than before. There is remains a large, primarily unchanged back-yard ‘lowinput low output, poultry system commonly found in poor rural households irrespective of
where they are. In addition, there is now also a significant small-medium scale poultry sector
benefiting from many of the technologies and improved breeding stock, feeds etc to meet
local and national demand. Given access to a reliable supply of inputs and a market for their
products these can be successful. They will however become increasing squeezed when
having to meet increasingly stringent and enforceable health and safety regulations but
without the benefits of the economies of scale and vertical integration of the industrial sector.
There is expected to be an increasing dichotomy between the industry and the back-yard
systems at the expense of the middle-ground. As long as there is rural poverty, back-yard
systems will survive and continue to make an important socio-economic contribution.
The growth of the poultry sector has not been without its negative consequences. A
number of serious or potentially serious human diseases have their origins in the poultry
sector. Intensive production systems, high poultry densities often associated with high human
population densities and exacerbated in many cases by ineffective standards and non-existent
biosecurity, have proven to be an ideal breeding ground for emerging new diseases. There are
also issues of pollution, greenhouse gas emission, erosion of biodiversity and the food versus
feed grain issues that remain largely unresolved.
FAO originally supported a major field programme providing direct assistance to
poultry producers in the developing world. As the technical capacity in many countries
continues to grow, as well as the parallel growth in the scale, scope and expertise of the NGO
operations, FAO’s traditional field programme has declined. With a well established
programme of decentralization, direct support to member countries is the responsibility of
country offices and technical teams based at the sub-regional level. The exception is the
emergency and rehabilitation programme which continues to grow and coordination remains
at headquarter level.
168
Aust. Poult. Sci. Symp. 2010...21
Technical divisions at FAO headquarters have taken on a more informative and
analytical role. AGA anchors its livestock programme around the three international public
goods: poverty reduction; public health and sustainable resource use. This has proved an
effective way to examine some the major issues and challenges affecting the poultry sector.
As long as there are rural poor, back-yard ‘low-input - low-output’ systems will continue to
make an important socio-economic contribution to improving livelihoods. Major epizootic
and zoonotic diseases are associated with poultry production, and the poultry sector has
historically been a major source of emerging diseases. Given the volume of fresh and highly
perishable poultry products traded each year, food safety is not surprisingly an issue,
especially in countries and markets where standards are either not available or enforced. The
impact of the poultry sector on the environment includes potential pollution, direct and
indirect greenhouse gas emissions and the food-feed issue. These are all areas in which, to
some extent, AGA has on-going activities, for example: poultry production, animal health,
policy support, management of animal genetic resources, animal welfare and livestock and the
environment.
The poultry sector will continue to adapt and change. FAO will continue to have an
important role in analyzing, guiding, advising, informing and warning to ensure the sector
makes a safe, secure and equitable contribution to human and economic development. The
2009 FAO State of the World’s Agriculture (SOFA) is entitled Livestock in the Balance (to be
published in early 2010) examines the major issues associated with the livestock sector
including poultry. As the title suggests, it reviews its positive contribution to human
development as well as the more negative consequences and, importantly, how these can be
mitigated. The 2009 SOFA is a good example of the way in which AGA now works to
support the livestock sector.
REFERENCES
Dolberg F (2008) Poultry production for livelihood improvement and poverty alleviation.
Proceedings of the International Conference Poultry in the Twenty-first Century:
Avian Influenza and Beyond. FAO, Rome.
FAO (2007) The State of the World’s Animal Genetic Resources for Food and Agriculture,
Rome.
FAO (2008) The Global Strategy for the Prevention and control of H5N1 Highly Pathogenic
Avian Influenza, Rome.
FAO (2009) FAOAIDE news Situation Update 62, Rome.
Gerber P, Opio C, Steinfeld H (2008). Poultry production and the environment - a review.
Proceedings of the International Conference Poultry in the Twenty-first Century:
Avian Influenza and Beyond. FAO, Rome.
Hinrichs J, Steinfeld S (2008). Feed availability inducing structural change in the poultry
sector. Proceedings of the International Conference Poultry in the Twenty-first
Century Avian Influenza and Beyond. FAO, Rome.
Kiiholma J (2008). Food safety concerns in the poultry sector of developing countries. In
Thieme O & Pilling D, eds. Proceedings of the International Conference Poultry in
the Twenty-first Century: avian influenza and beyond. FAO, Rome.
McLeod A (2008). Social impacts of structural change in the poultry sector. Proceedings of
the International Conference Poultry in the Twenty-first Century: Avian Influenza and
Beyond. FAO, Rome.
Pym RAE, Farrell DJ, Jackson CAW, Glatz PC, Mulder AW (2009) Technological Change
and its Impact on Poultry Development – a review. Unpublished report. FAO, Rome.
169
Aust. Poult. Sci. Symp. 2010...21
Sims LD (2008). Risks associated with poultry production systems. Proceedings of the
International Conference Poultry in the Twenty-first Century Avian Influenza and
Beyond. FAO, Rome.
Sonaiya F (2008). Smallholder family poultry as a tool to initiate rural development.
.Proceedings of the International Conference Poultry in the Twenty-first Century:
Avian Influenza and Beyond. FAO, Rome.
Upton M (2008). Scale and structure of the poultry sector and factors inducing change:
intercountry differences and expected trends. Proceedings of the International
Conference Poultry in the Twenty-first Century: Avian Influenza and Beyond. FAO,
Rome.
Vinod A, Arindam S (2008). Scope and scale for small-scale poultry production in developing
countries. Proceedings of the International Conference Poultry in the Twenty-first
Century: Avian Influenza and Beyond. FAO, Rome.
170
Aust. Poult. Sci. Symp. 2010...21
IMPROVED VILLAGE POULTRY PRODUCTION IN THE SOLOMON ISLANDS
R. PARKER 1
Summary
Village poultry keeping is a crucial activity in the Solomon Islands because over 80% of the
500,000 plus population lives in a rural subsistence existence generally without supplementary
wages, pensions or other income. Providing the family’s daily food requirement is often
impossible, particularly as a balanced diet. Most rural households have a food garden but despite
the lush tropical environment more than 30% of the children are malnourished. Protein crops in
the gardens are limited and the population generally is lacking protein. Traditional wildlife
sources of protein are depleted either from past hunting or environmental damage by logging or
commercial fishing. Increased consumption of chicken meat and eggs through improved village
chicken keeping is a quick and simple solution to these problems as well as a relief response to
disasters like famine, storm, tsunami and earthquake. It has been recognised from the Prime
Minister down that improved village chicken keeping is one of the two key development
activities which are necessary for the country. Unfortunately the Agriculture Department does
not yet have the capacity to provide the training and support needs for this activity. The author’s
training and development program, Kai Kokorako Perma-Poultry is attempting to meet the needs
of the whole country.
I. INTRODUCTION
Many rural villages have a few chickens wandering freely without any formal care or attention.
As breeding is not structured there is often and imbalance of roosters over hens causing uncertain
fertility and breeding results. Eggs are laid in the bush and sitting hens or their eggs are often lost
to predators. A broody hen can hatch up to 15 chicks however she is often left with one or two
chickens to raise to maturity because of lack of feeding and is therefore left out of production for
a longer period than necessary. The use of commercial feeds is usually cost prohibitive and
distribution networks of this feed is often unreliable or difficult.
II. IMPROVEMENT PROPOSAL
Commencing in 1994 it was recognised by the author that the simple backyard poultry keeping
hobby practiced by him and many other Australians was an ideal improvement method for
Solomon Island households. He then commenced writing a training manual to suit Solomon
Island conditions and needs. It was decided to address four main husbandry issues: (i) breeding,
(ii) feeding, (iii) housing as well as (iv) general care and management.
The training principles are presented as “Kai Kokorako Perma-Poultry” because Kai
Kokorako means “eat chicken” in Pidgin English and Perma-Poultry refers to the permaculture
and sustainable nature of the husbandry methods. Village chicken housing is built from the same
1
PO Box 242. Tenterfield, NSW 2372. Australia
171
Aust. Poult. Sci. Symp. 2010...21
climate friendly bush materials as the peoples’ own houses and the chicken feed is provided in
the form of a balanced diet from the peoples’ own food gardens. Each chicken flock is stocked
with a crossbred mixture of existing village chickens, hybrid layers and broilers which have
naturally become a hardy, self reliant local breed able to survive the difficult village lifestyle.
This improvement program is meeting the development needs in two major ways.
III. TRAINING
The training manual was first published as a general village training book in 2004 in association
with the Rural Youth Livelihoods Project resulting from Aid responses to the needs of displaced
youth from the Ethic Conflict. Some 250 rural youth and villagers benefited from exposure to
training workshops conducted using this manual with the result of several substantial village
farms being successfully established.
Later it was further recognised that many more competent trainers were needed if the
whole country was to benefit. The manual was therefore rewritten and republished in 2007 in the
format of “Train the Trainer “ and made available online through the nine Distance Learning
Centres (DLC) across the country as well as by CD Rom and hardcopy.
Training workshops are currently delivered both onsite and online by the author through
the various DLCs with the assistance of local tutors seconded from the Department of
Agriculture and various schools or colleges as well as Rural Training Centres.
A further development program focusing on the construction of small chicken breeding
farms alongside each DLC has commenced so that practical experience can also be offered at the
time of the written training. Many of these DLCs are constructed within close proximity to
schools or colleges so these practical farms can also complement each school’s curriculum and
as many are boarding schools, the larger farms also contribute to the food needs of the students.
IV. STRUCTURED BREEDING
The existing village chicken is usually a small bird with a mixture of bloodlines including
imported pure breeds, hybrid layer and broiler. The small size of the village chicken results not
only from the semi-feral lifestyle but also from an almost accidental infusion of bloodlines from
the wild chicken of St Cruz Island. This closely resembles the red jungle fowl in appearance and
size. These village chickens are hardy, good scavengers and because of their various mixed
bloodlines respond quickly to better feeding and care.
No hatchery currently exists in the country specifically for village chickens so there is
limited stock available to meet the peoples’ needs. Only day old layer and broiler chickens are
commercially available however they need commercial feeds to survive and produce well. By
crossing the hybrid layers and broilers with village and wild chickens their progeny are more
able to cope with garden based feed supplies. There is a further need for larger regional farms to
be built to produce suitable replacement stock using natural methods because of lack of access to
electricity and cost- prohibitive incubators and brooders.
The increased importation restrictions imposed since the Bird Flu outbreaks around the
world have made it impossible to import the necessary improved breeds of chicken suited to
village life. This has meant that any breeding improvement activities within the country need to
be carried out with what is already available.
172
Aust. Poult. Sci. Symp. 2010...21
Fortunately on the small far northern atoll of Ontong Java the chicken population is
dominated by a previous importation of valuable Naked Neck chickens. Two years ago a
breeding program was put in place to increase the numbers of this valuable breed so that they can
be conserved and be distributed more widely across the country to various breeding upgrade
programs. Early in 2009 this breeding project was placed under threat by flood tides so
conservation and wider distribution was made even more urgent. Whilst Ontong Java was not
within the 2007 tsunami affected area of the Solomons it is still constantly at risk because it is
low lying and unprotected. Additional breeding improvement projects are planned for wild
chicken conservation and captive breeding to provide the necessary breeding upgrade stock for
the local chicken hatcheries.
V. CONCLUSION
Whilst funding for these various improvement and developmental activities is limited it is
believed that by combining the continuing training program and establishment of breeding
centres across the country a huge improvement can not only be made to human nutrition and
health but also unique wildlife can be conserved. The planned development of both a central
hatchery as well as several regional hatcheries will help provide the necessary breeding and
replacement stock to ensure success of the whole program. It is also recognised that the more
successful breeding farms, hatcheries and training centres can provide employment opportunities
and small incomes for necessary personal needs like school fees, kerosene for lighting and
supplementary food items or developmental needs like rural electrification and water supplies.
173
Aust. Poult. Sci. Symp. 2010...21
HEAVY METAL CONTAMINATION IN MINERAL SOURCES FOR MONOGASTRIC
FEED IN ASIA PACIFIC
T. JARMAN 1, A. FRIO 2, A. LEARY1, A. KOCHER 3, S. FIKE 4 and B.TIMMONS4
Summary
A survey of mineral supplements for pig and poultry feeds was conducted in Asia-Pacific to
determine the presence and extent of heavy metal contamination across the region. Types of
materials sampled included inorganic mineral premixes, zinc oxide (ZnO), zinc sulfate
(ZnSO4), copper sulfate (CuSO4), ferrous sulfate (FeSO4), manganous oxide (MnO),
manganous sulfate (MnSO4), sodium selenite, chelated minerals/premixes, and complete
feeds with inorganic minerals. Countries surveyed were Australia, China, India, Malaysia,
Pakistan, Philippines, Taiwan, Thailand and Vietnam. It was found that of 275 samples
submitted, 17% were found to be contaminated at least one heavy metal in excess of EU
limits was present.
I. INTRODUCTION
Mining is the most common source of inorganic minerals used in the feed industry. Without
careful processing and filtration, heavy metal contamination of inorganic minerals can find its
way into the feed chain. Detection of the contamination is dependent on adequate monitoring
systems. Too often, the contamination is not found until the inorganic mineral is already
made into pre-mixtures or complete feeds.
The European Union has implemented a Rapid Alert System for Food and Feed
(RASFF) to detect and alert member countries when risk products have made it into the
market place. In 2007, China was the leading source of alerts for contaminates such as heavy
metals (RASFF, 2007). A Chinese origin contamination scandal occurred in Australia in 2008.
Zinc oxide containing high levels of lead found its way into the Australian feed industry and
excessive lead levels were detected in the livers and kidneys of pigs fed the zinc oxide
(Spragg, 2008).
High cadmium contamination in zinc sulfate lead to infertile eggs being laid by
breeder hens in South Africa in 2007 (Johns, 2007). The young chickens had severely
reduced growth. Organic minerals are also at risk of contamination. The need for a wide and
frequent screening system for all minerals is required by suppliers to ensure excessive levels
are detected before they enter the feed chain.
The effects of heavy metal exposure in poultry are not widely reported. Lysenko
(2005) found that adding 100 ppm of lead to a broiler diet rendered all edible parts of the
chicken unfit for human consumption. Exposing laying hens to lead in the diet will result in
increases in lead levels in the tissues, organs and egg yolk (Trampel et al., 2003).
Human effects of heavy metal exposure are varied. Detrimental effects to a number of
vital systems and organs have been seen. Cadmium can have serious negative effects on the
renal and respiratory systems in humans and is not easily eliminated (Palachy et al.,
1998). Järup (2003) reports that long-term exposure to arsenic in drinking-water has been
associated with increased risks of skin cancer.
1
Alltech Asia-Pacific Bioscience Centre, 113, Thailand Science Park, Pathumthani, Bangkok, 12120
Alltech Philippines, Muntinlupa City, Philippines
3
Alltech Biotechnology Pty. Ltd. Dandenong South VIC Australia
4
Alltech Center for Animal Nutrigenomics and Applied Animal Nutrition, Nicholasville, KY, USA
2
174
Aust. Poult. Sci. Symp. 2010...21
It was theorised by the authors that the occurrence of heavy metal contamination in
the Asia-Pacific region was more wide-spread than previously thought. Therefore, the
objective of the current survey was to gather and analyise a range of mineral types and
sources from Asia-Pacific countries and evaluate for heavy metals.
II. MATERIAL AND METHODS
A total of 275 100 g samples were collected in the southern hemisphere autumn of 2009.
Types of materials sampled included inorganic mineral premixes, zinc oxide (ZnO), zinc
sulfate (ZnSO4), copper sulfate (CuSO4), ferrous sulfate (FeSO4), manganous oxide (MnO),
manganous sulfate (MnSO4), sodium selenite, chelated minerals/premixes, and complete
feeds with inorganic minerals. Countries surveyed were Australia, China, India, Malaysia,
Pakistan, Philippines, Taiwan, Thailand and Vietnam. All samples were analyzed for lead,
arsenic, mercury, and cadmium using inductively coupled plasma mass spectrometry (ICPMS) at the Alltech Center for Animal Nutrigenomics and Applied Animal Nutrition in
Kentucky. Metal concentrations were compared against EU contamination limits for animal
feed. Maximum allowable EU limits for lead, arsenic, mercury and cadmium are 100 ppm, 15
ppm, 0.05 ppm and 10 ppm respectively for feed additives and 30 ppm, 50 ppm, two ppm and
0.5 ppm for complete feed.
III. RESULTS
50
45
40
35
30
25
20
15
10
5
ra
ll
ve
O
m
na
Vi
et
d
Th
ai
la
n
Ta
iw
an
s
Ph
ili
pp
in
e
an
is
t
Pa
k
ia
M
al
ay
s
di
a
In
Au
st
ra
C
hi
na
0
lia
% Contamination with at least 1 Heavy
Metal
By country, the percentage of feed material samples with levels of at least one heavy metal
above allowable EU limits is as follows: Australia (27%), China (3%), India (43%), Malaysia
(32%), Pakistan (33%), Philippines (18%), Taiwan (11%), Thailand (11%), and Vietnam
(12%). Overall, 17% of samples analysed, tested high for at least heavy metal in the AsiaPacific region.
Country
Figure 1
Percentage of samples contaminated with at least 1 heavy metal by country.
175
Aust. Poult. Sci. Symp. 2010...21
Of 25 Poultry pre-mixes containing inorganic minerals, 48% were found
contaminated with at least 1 heavy metal. The highest level of lead, arsenic, mercury and
cadmium was 697.12 ppm, 90.46 ppm, 0.14 ppm and 111.01 ppm respectively. All are above
EU allowable limits for feed materials.
Individual minerals were analysed and found that the following were contaminated
with at least one heavy metal in excess of EU limits: ZnO (12 samples) 25%; ZnSO4 (12
samples) 25%; CuSO4 (21 samples) 19%; FeSO4 (13 samples) 15%; MnO and MnSO4 (15
samples) 20%; sodium selenite (six samples) 60% and Non-Alltech organic mineral pre-mix
(29 samples) 10%. The highest level of lead, arsenic, mercury and cadmium was 632.19 ppm,
659.52 ppm, 1.63 ppm and 3982.67 ppm respectively. All are above EU allowable limits for
feed materials.
Of 30 complete feeds containing inorganic minerals, 7% was found contaminated
with at least 1 heavy metal. The highest level of lead, arsenic and cadmium was 277 ppm,
8.05 ppm, and 0.84 ppm respectively. All are above EU allowable limits for feed materials.
Table 1.
Heavy metal levels of inorganic poultry pre-mixes (n = 25).
Heavy
Maximum
Samples over
metal
limit (ppm)
max. limit (%)
Lead
< 100
8
Arsenic
< 15
20
Mercury
< 0.05
28
Cadmium
< 10
8
Contaminated with at least 1 heavy metal: 48%
Table 2.
Item
ZnSO4
ZnO
CuSO4
FeSO4
MnO/MnSO4
Na2SeO3
Mineral px.
Table 3.
Heavy
metal
Lead
Arsenic
Mercury
Cadmium
Lowest
0.21
0.09
ND
0.06
Levels (ppm)
Highest
697.12
90.46
0.14
111.01
Mean
49.73
12.3
0.02
0.08
Heavy metal levels of individual inorganic minerals and Non-Alltech organic
minerals.
Lead (ppm)
Max.
Mean
50.93
5.58
632.19 83.96
20.51
2.73
0.09
0.62
68.42
11.9
0.81
0.44
87.86
6.05
Arsenic (ppm)
Max.
Mean
149.36 19.02
17.94
2.75
659.52 38.74
1.04
0.53
26.99
4.44
65.55
15.75
65.92
3.78
Mercury (ppm)
Max.
Mean
ND
ND
0.07
2.42
0.07
0.07
0.51
0.05
1.63
0.11
1.21
0.24
0.22
0.01
Cadmium (ppm)
Max.
Mean
3982.6 369.6
12.02
2.42
1.33
0.14
75.92
6.95
19.43
1.6
0.08
0.02
12.35
1.14
Sample
numbers
12
12
21
13
15
6
29
Heavy metal levels of complete feeds
Maximum limit
(ppm)
< 30
< 50
<2
< 0.5
Lowest
0.06
0.36
ND
0.02
176
Levels (ppm)
Highest
277
8.05
ND
0.84
Mean
6.69
1.56
ND
0.06
Aust. Poult. Sci. Symp. 2010...21
IV. CONCLUSION
In conclusion, contamination with at least one heavy metal in excess of EU limits was present
in feed materials from all Asia-Pacific countries sampled. Of the 275 samples analysed, 17%
tested high for at least one heavy metal. Excessive exposure to heavy metals in the feed will
have a negative affect on animal health and performance.
REFERENCES
European Commission (2007) The Rapid Alert System for Food and Feed Annual Report, 170.
Järup L (2003) British Medical Bulletin 68, 167-182.
Johns L (2007) The Star - Food safety needs to be taken seriously 4.
Lysenko M (2005) Proceedings, XVIIth European Symposium on the Quality of Poultry
Meat, 333-335.
Plachy J, Jami AT, Coleman RR (1998) Cadmium 15 1-6.
Spragg, J (2008) Stock Feed Manufacturers’ Council of Australia Members Notice Zinc
Oxide and Lead Contamination Update.
Trampel DW, Imerman PM, Carson TL, Kinker JA, Ensley SM (2003)
Journal of Veterinary Diagnostic Investigation 15, 418–422.
177
Aust. Poult. Sci. Symp. 2010...21
DUCK PRODUCTION IN AUSTRALIA
P. R. BROWN 1
I. INTRODUCTION
Up until the 1970’s duck production in Australia was confined mostly to small scale
operations producing ducks from stock that took up to 12 to 14 weeks to grow to a
marketable size. Over the past three decades the industry has undergone rapid growth due to
increasing demand for product and this has led to the emergence of two large scale vertically
integrated businesses, Pepe’s Ducks based at South Windsor, New South Wales and Luv a
Duck based at Nhill, Victoria. This growth has seen one business expand from a backyard
situation with twenty two ducks to an annual production exceeding 3.5 million ducks. Overall
production in Australia is around 8 million ducks per year. Duck production in Australia
today is valued at about $100 million per annum. The domestic market is based mostly on the
supply of ducks to Asian communities throughout Australia. Approximately 75% of all ducks
processed are supplied to Chinese BBQ shops, Asian restaurants and butchers. The remaining
25% is supplied to fine dining restaurants, delicatessens, major retail outlets and export
markets. Over the past five years industry sales have grown by about 60% and future growth
appears to be exceptionally good throughout all market sectors.
II. DUCK BREED AND PERFORMANCE
The Pekin duck is the breed of choice for duck meat production as it can be reproduced all
year round and is a large sized meaty duck that performs well in commercial farming
operations under varying conditions. The importation of genetic stock in recent times from
Grimaud in France by Pepe’s Ducks and from Cherry Valley in the United Kingdom by Luv
a Duck has significantly improved production performances and meat yields. The
performance for Pekin ducks varies depending on the genetic stock but in general the
following standards apply for duck production in Australia. Breeding ducks commence egg
production at 24 weeks of age and their peak egg production is at 32 weeks with 88 to 92%
production. The length of their laying period is 40 to 45 weeks with a production of 200 to
220 eggs with a hatchability of 80-85%. Meat ducks are processed at 42 days of age at a 2.85
kg live-weight with a feed conversion ratio of 2.20 and water consumption of 18.5 litres. The
mortality rate is in the order of 3.0%.
Throughout other parts of the world it has been more common practice to grow ducks
to a processing age of 47 days and a live-weight of 3.40 kg. However, in the past year and a
half financial pressures have required that the processing age be reduced to 42 days and with
a live-weight of 3.0 kg but still maintain the same meat yield percentages as of a 47 day duck.
With increasing prices for cereal grain and other essential raw ingredients the lowering of the
age and weight is a solution to reducing feed costs which represents 58% of the cost of
producing a duck. To meet this challenge duck breeders in the northern hemisphere have bred
stock to achieve these requirements. To keep up with the rest of the world, supply of this new
genetic stock will be imported to Australia starting as of 2010.
1
Pepe’s Ducks Pty Ltd. 17 Walker Street, South Windsor NSW 2756.
178
Aust. Poult. Sci. Symp. 2010...21
III. BROODING AND GROWING DUCKS
The brooding and growing of ducks is mostly conducted in deep litter conventional shedding
comprising typically of insulated roofing, enclosed end walls, wired side walls with
adjustable curtains, nipple drinkers, automatic pan feeders, hot spot gas fired brooders,
lighting, fans and foggers. Controlled environment shedding for duck production in Australia
is still in its infancy. Brooding and growing ducks is very similar to other poultry species
with the exception of water. Since ducks are a waterfowl species they require a plentiful
supply of water not only for consumption but also for preening. Contrary to common belief
ducks are very clean birds and perform best when they can rest on fresh dry litter material.
The fact that ducks require more water and have web feet makes litter management a critical
component of husbandry. It is important that fresh litter material is added regularly
throughout the growing cycle to maintain dry fresh litter conditions. Litter requirements also
impact directly on stocking densities which can vary from 5 to 6 ducks per square metre in
winter and 4 to 5 ducks per square metre in summer. Meat duck production is mostly
conducted on company operated farms and independent operated farms with a growing
agreement to supply a processor.
IV. NUTRITION
The feeding programme for meat ducks is usually a combination of starter and grower feeds.
For the first two weeks of production ducks are commonly feed a coarse starter crumble with
a protein level of 21% to 22% and energy level of 12.2MJ/kg. After two weeks of age a
grower pellet consisting of 17% to 18% protein and energy level of 12.7MJ/kg is fed to the
ducks up until the time of processing. Breeding ducks are fed a program of starter, grower
and developer feed during the rearing phase and breeder feed during egg production. Crumble
and pellet durability is of critical concern for duck production. Ducks tend to shovel feed into
the bill rather than peck at feed. Due to this action, ducks prefer solid crumbles and pellets
and avoid fines wherever possible. It is therefore important that the integrity of the crumble
and pellet structure is maintained to minimise fines and production losses. Wheat is the
preferred grain to feed to ducks due to better performances achieved with this grain. Wheat
also has good glutinous characteristics and this can assist with feed binding and can minimise
the percentage of fines in the finished product.
V. PROCESSING
Processing ducks is not as straight forward as for other species of poultry due to the thicker
layering of feathering that ducks produce. Removal of feathering is a challenge and requires a
three stage process of scalding, plucking and waxing. The presence of pin feathers on ducks
can cause feather removal issues and this is a worldwide problem within the industry. In the
past it was common practice for duck producers to process ducks using modified equipment
that was designed and manufactured for processing broiler chickens. Today equipment is
designed and built specifically for the purpose of processing ducks. The scalder, plucker, wax
system, eviscerator and packing equipment are controlled individually by a Program Logical
Control (PLC) system which can be programmed to set parameters for age, weight and strain
of duck in order to maximise processing efficiency. This purpose designed and built
equipment can be found today in some Australian duck processing plants and these plants are
equal to any in the world. The meat, fat, skin and bone percentages for a 42 day processed
duck with head off and minus wings varies depending on the genetic stock. In general, meat
comprises 23-24% of the carcass, fat and skin 33-34% and bone 31-32%. The fat content is
179
Aust. Poult. Sci. Symp. 2010...21
an essential component of the duck and this gives the meat a tender and favoured appeal
which is most important to the Asian community.
VI. HEALTH
Ducks are very robust and are not subjected to any significant viral or bacterial related
diseases in Australia and therefore do not require routine vaccination or treatment
programmes. The most severe disease to affect ducks in Australia is Rimerella (Pasteurella)
anatipestifer. Conditions such as ascites, leg weakness and other physical conditions are of
minor consequence and concern.
VII. BIOSECURITY
The two major duck producers are committed to biosecurity and have implemented standards
and procedures that comply with the Australian Government Department of Agriculture,
Fisheries and Forestry National Farm Biosecurity for poultry production. Maintaining these
biosecurity standards is important to prevent diseases and pests from infecting poultry
premises and may be the mechanism that continues to protect the poultry industry from the
threat of imported poultry products.
VIII. RESEARCH AND DEVELOPMENT
The Australian duck industry has been neglected in the past in respect to domestic based
research and development work. The industry has relied heavily on research and technical
information developed primarily for the production of ducks in the northern hemisphere and
has adapted this information to Australian conditions. This situation has turned around in the
last few years with a partnership forged with a major duck producer, Rural Industries
Research and Development Corporation and Sydney University. This has led to extensive
trial work being conducted to establish the strain effect on growth rate, feed consumption,
water usage, processing and cooking. This partnership is set to continue with further trials to
be conducted to investigate the effects of stocking density, transportation, summer heat and
behaviour in relation to performance.
IX. AUSTRALIAN DUCK MEAT ASSOCIATION
The Australian Duck Meat Association (ADMA) was formed and incorporated in 2008 as an
initiative of the two major duck producers. The association was formed for the following
objectives:
(i) To provide a forum whereby duck meat processors can be represented by a
common association for the long term being of the industry. (ii) To offer protection of
financial interests for members and to ensure a reasonable financial return as the industry
grows. (iii) To promote duck meat as an alternative food base and gain a greater acceptance
by consumers. (iv) To establish and maintain compliant growing and processing standards to
further enhance the industry. (v) To work closely with state and federal bodies and other
representative bodies to maintain a high standard of operation for the benefit of the entire
industry. (vi) To maintain and pursue practices that will help ensure a disease free status
giving consumers confidence in a wholesome meat product. (vii) To develop and monitor
strict biosecurity standards resulting in optimum disease control and to help reduce the
likelihood of serious disease spreading.
In 2008, the Australian Duck Meat Association became a member of Animal Health
Australia (AHA). Since becoming a member, the ADMA has been actively involved in AHA
matters participating at forums, workshops and staff training events.
180
Aust. Poult. Sci. Symp. 2010...21
X. CHALLENGES AND THE FUTURE OF THE DUCK INDUSTRY
There are a number of challenges facing the industry, which must be countered in the future.
These include (i) sourcing long term adequate supplies of suitable litter material at affordable
prices or the development of viable alternative products or systems. (ii) The threat of duck
meat products being imported into Australia from countries such as China, Thailand and
Vietnam. (iii) Long term provision of a Quarantine facility for the importation of genetic
stock. (iv) Increasing prices for cereal grains and other raw ingredients for the production of
feed. (v) Development of value added products and introduction of new product lines to meet
the requirements of customers. (vi) Urban encroachment on poultry farming areas. (vii)
Recruitment of staff.
The demand for duck meat continues to grow from year to year in Australia and the
industry is also aware that to the north of the country lays the largest number of duck
consumers in the world, that being Asia. China, Thailand, Vietnam, Taiwan, Korea, Hong
Kong and Malaysia have the potential to develop into large export markets for Australian
duck meat products. As much as there is a huge amount of duck grown and processed in Asia
their greatest threat is from diseases especially highly pathogenic avian influenza. With
modern growing and processing facilities, world class genetic stock, high standards of
biosecurity and focus on animal welfare this will hold the Australian industry in good stead to
supply duck product to the world’s largest markets.
181
Aust. Poult. Sci. Symp. 2010...21
THE EFFECTS OF STRAIN AND SEASON ON THE PERFORMANCE OF COMMERCIAL
DUCKS UNDER AUSTRALIAN CONDITIONS
J.A. DOWNING 1 and W. TAYLOR1
Summary
The Australian duck industry has a very specific market requirement, this being for a 2.85 kg
bird at 6 weeks of age. The strains of Pekin duck presently used in Australia, the Cherry Valley
and Grimaud Frères, have different growth characteristics but both have difficultly meeting this
target weight especially in summer. It was considered that by crossing these two strains, hybrid
vigour might allow advantages to be gained in growth performance. The present study
investigated the performance of the two main strains of Pekin ducks and their reciprocal crosses
grown to 6 weeks of age in summer and winter. Ducks were reared following industry practices.
The strains and their crosses were bred by PE’S Ducks Pty Ltd, and reared in single sex groups
or as mixed sex groups. In summer only one strain reached market weight by 41 days of age. In
winter all strains reached market weight by 41 days but the FCR was higher in winter than
summer. Males grew to heavier weights than females in both summer and winter but there was
no advantage gained by rearing ducks as single sex groups.
I. INTRODUCTION
Because it is a relatively new industry, the amount of information specific to duck production
under Australian conditions is limited. At present, two different strains of Pekin duck are used by
the Australian industry, the Grimaud Frères (GF) and the Cherry Valley (CV). Both have distinct
growth characteristics but because they have been developed in Europe to meet needs of the
local market, individually they fail in some respects to meet the very stringent requirements of
the Australian market. In Australia the current market specifications are for a bird grown to 2.85
kg at 6 weeks of age to meet the needs of the whole bird restaurant trade. Another issue of
industry concern is the late breast muscle development in Pekin ducks. So getting a strain that
grows rapidly to market weight with a high yield of breast muscle is a major contradiction faced
by local producers. The GF grows to larger size than the CV but matures later and is considered
to produce meat of lower eating quality. One consideration given to obtaining a better breeding
outcome to meet local needs is to cross the GF and CV strains and achieve improved
performance from hybrid vigour.
A primary objective of the present studies was is to evaluate the performance of the CV
and GF strains and the reciprocal crosses of these strains reared under Australian summer and
winter conditions. A further objective was to evaluate the performance of different sexes and the
affect of rearing ducks as single or mixed sexes. One study was carried out in February/March
and the other in June/July 2007.
1
Faculty of Veterinary Science, University of Sydney, Camden NSW, 2570.
182
Aust. Poult. Sci. Symp. 2010...21
II. METHODS
Studies using the same experimental design were undertaken in summer (Feb-Mar) and in winter
(Jun-Jul). Birds were reared in a tunnel ventilated shed with 48 floor pens (1.5m x 3.0m) with
four replicate pens allocated to each treatment. The ducks were raised on deep litter, consisting
of wood shavings. The four strains, GF, CV and the reciprocal crosses, GF male x CV female
and CV male X GF female, were breed by the industry partners, PEPE’S Ducks Pty Ltd. Eggs
were incubated at their commercial hatchery and at hatch ducklings were vent sexed and
transported as day olds to The University of Sydney, Camden. On arrival 36 ducklings of a
specific strain were allocated at random to the appropriate treatment pens as single sex groups
(male or female) or mixed sex groups (equal numbers of males and females). On day six wing
tags were inserted into the right wing of all birds. Birds were fed a commercial starter crumble
diet on days 1-14 formulated to supply 12.14 MJ/kg ME and 22% protein, and pelleted grower
diet from days 15-41 formulated to provide 12.77 MJ/kg ME and 19% protein. The ducks were
provided with feed ad libitum. Each pen had its own water supply which ducks accessed via a
row of 4 nipple drinkers. Birds were weighed individually at the end of each week, except in
week 6 when they were weighed at day 41. Both feed intake and water intake were determined at
the end of each week on a pen basis. Throughout the study birds were removed from the pens for
carcass measurements and a metabolism study. At the start of week six the bird density was 22
birds per pen in summer and 24 birds per pen in winter.
Data were analysed using the REML linear mixed model function of Genstat® 11th
edition. Data was loge transformed where necessary. The fixed model included the effects of
strain, sex, pen-sex and week and the random model included the effects of block, pen and bird
identification. Significance testing of fixed effects was conducted using Wald chi-square tests
with a significance threshold of P < 0.05. Any non-significant interactions were removed from
the model and for significant effects the least significant difference (LSD) procedure was used to
make pair-wise comparisons of means.
III. RESULTS
For reasons of ‘commercial-in-confidence’ the strains will be identified as A, B, A X B and B x
A. The final mean (± SEM) six week liveweight (LW), FCR and water to feed ratio for the
effects of strain and sex are given in table 1. In summer and winter there were highly significant
effects of strain, sex and pen-sex on LW but these effects changed over time, as indicated by the
significant interactions with week (all P< 0.001). Mean LW were lower in summer than winter.
At the end of the six week production period, ducks of strain B (P < 0.05) were heavier than
other strains in both seasons, followed by strain A X B (P < 0.05) being heavier than strains A
and B x A again in both seasons. In summer strain A and A X B had similar weights but in
winter strain B X A (P < 0.05) was heavier than strain A. In summer, males were heavier than
females at all weeks (P < 0.05) and similarly in winter, except for week 1, males were again
heavier than females (P < 0.05). The differential between males and females was similar for both
seasons, approximately 200g. In both seasons the pen-sex x week interaction (P < 0.001) was
significant but on no individual week were the differences significant suggesting the effect is
questionable.
183
Aust. Poult. Sci. Symp. 2010...21
Table 1.
Strain
The mean (± SEM) six week liveweight (LW), FCR and water to feed ratio for
two strains (A and B) of commercial Pekin ducks and their reciprocal crosses (A
X B and B x A). For these measures the effect of sex is also given.
Six week
Liveweight
(g)
Strain
A
2705
± 33c
2844
± 34b
2963
± 39a
2724
± 33c
AXB
B
BXA
Sex
Female
Male
2708
± 34b
2910
± 32a
FCR
Water to
feed ratio
Six week
Liveweight
(g)
Summer
1.95
3.34
± 0.02b
± 0.05a
2.00
3.43
± 0.02a
± 0.05a
1.96
3.33
± 0.02ab
± 0.05a
1.89
3.23
± 0.02c
± 0.05b
Summer
2.01
3.36
± 0.01a
± 0.04
1.89
3.29
b
± 0.01
± 0.04
FCR
Water to
feed ratio
Winter
2847
± 20d
3106
± 22b
3262
± 23a
2951
± 21c
2942
± 18b
3137
± 19a
2.09
± 0.01
2.10
± 0.01
2.10
± 0.01
2.06
± 0.01
Winter
2.14
± 0.01a
2.03
± 0.01b
2.70
± 0.05
2.75
± 0.05
2.68
± 0.05
2.70
± 0.05
2.76
± 0.04a
2.68
± 0.04b
Within columns for strain and sex values with different superscripts are significantly different.
In summer, the mean (± SEM) LW at six weeks for birds reared in single pens (2821 ± 34
g) and mixed pens (2793 ± 31 g) was not different. In winter, the mean (± SEM) LW of birds
reared in single pens (3035 ± 21 g) and mixed pens (3038 ± 18 g) and again was not different. In
summer, strain had significant effect on FCR (P < 0.001). Strain B X A had a better FCR than
other strains (P < 0.05). Strain A had a lower FCR than strain A X B (P < 0.05) with the FCR for
strains B and A X B not different. In winter strain had no effect on the FCR (P=0.18). In both
summer and winter (P < 0.001) sex had a significant effect of FCR with males in both seasons
having a better FCR than females (P < 0.05). Strain influenced water to feed ratio in summer
(P=0.02) but not in winter (P=0.23). In winter, sex (P=0.04) had significant effect on the water to
feed ratio with males having a lower ratio than females. However, in summer sex had no
influence on water to feed ration (P=0.33).
IV. DISCUSSION
In Australia the market specification for ducks is for a liveweight of 2.85 kg at slaughter.
Processing ducks at 2.85 kg limits the yield of breast muscle and this represents a major
limitation faced by the Australian meat duck industry. It was considered that by forming
reciprocal crosses from matings of the main strains, a production advantage might be gained
from hybrid vigour and that the crosses may better meet the needs of the Australian market.
However, the crosses provided no advantage above the parent strains. Wawro et al. (2004)
reported that crossbred ducks did not demonstrate heterosis for liveweight. While on the surface
it appears that rearing ducks in summer is more efficient with a lower FCR, it needs to be
remembered that in commercial practice the rearing age would need to be increased to get ducks
184
Aust. Poult. Sci. Symp. 2010...21
to correct market weight in summer and that this would negate any benefit in FCR and actually
increased production costs because slaughter age would be increased.
In the Australian meat duck industry it is common practice to raise ducks in mixed sex
pens. This is adopted because it is widely believed that the sexual dimorphism in Pekin ducks is
small until 6 weeks of age (Farhat and Chavez, 2000). The current study supports the present
practice, as little advantage is achieved by rearing ducks as single sex groups. Rearing ducks in
mixed sex pens avoids the necessity of vent sexing, which is both an invasive and expensive
procedure. The differences between sexes of the same strain, in the current study, ranged from
142 g for Strain A to 272 g for Strain A X B. Differences in LW between sexes increased with
age and the strains with the highest growth rates showed the greatest levels of sexual dimorphism
at six weeks. Therefore, sexual dimorphism could, in part, be responsible for the large variation
of body weights observed at Australian processing plants. Cross breeding of the present strains
of Pekin ducks available in Australia will not solve the dilemma facing the industry, this being
the poor breast muscle yield at the desired market weight of 2.85 kg. The Australian industry
needs to find either management, nutritional or genetic solutions to alter the ducks’ growth
allometry so breast muscle development occurs at an earlier age.
V. ACKNOWLEDGEMENTS
We acknowledge the financial support given by The Rural Industries Research and Development
Corporation (New Animal Industries) and the industry partners in this work, Pepe’s Ducks, Pty
Ltd. We also acknowledge the technical support given by staff at University of Sydney Poultry
Research Unit, Camden.
REFERENCES
Baeza E, Dessay C, Wacrenier N, Marche G, Listrat A (2002) British Poultry Science 43, 560568.
Farhat A, Chavez ER (2000) Poultry Science 79, 460-465.
Wawro K, Wilkiewicz-Wawro E, Kleczek K, Brzozowski W (2004). Archiv für Tierzucht 47,
287-299.
185
Aust. Poult. Sci. Symp. 2010...21
THE PERFORMANCE OF COMMERCIAL DUCKS FED DIFFERENT DIETARY PROTEIN
CONCENTRATIONS DURING THE FINISHER PHASE
J.A. DOWNING 1 and J. ISCHAN1
Summary
The objective of the present study was to investigate the dietary protein requirements of three
commercial duck strains needed to obtain best performance and growth under Australian summer
conditions. Three lines of Pekin duck, one Grimaud Frère line and two Cherry Valley lines were
fed a commercial starter (days 1-14) and grower (days 15-28) diets and then 4 diets containing
the same metabolisable energy content (12.5 MJ/kg) but differing in protein content (diet A,
16.8%; diet B; 16.1%; diet C, 15.4% and diet D 17.5%) from days 29-41. Individual ducks were
weight weekly and feed intake was measured weekly on a pen basis. No differences in liveweight gain (LG) were seen in weeks 1-4. In week 5, LG was lower for birds fed diets B and C
compared to those fed diets A and D (P < 0.05). In week 6, the effects were reversed, with LG
for birds fed diets B and C being greater than for birds fed diets A and D (P<0.05). Except for
week 1, the live-weight (LW) for strain X was less than for both strains Y and Z (P < 0.05). For
the first 4 weeks there was no difference in the LW for strains Y and Z. However, in weeks five
and six the LW for strain Y was greater than for strain Z (P < 0.05). The data indicate that
producers can reduce costs by introducing a low protein finisher diet into the feeding program of
commercial ducks.
I. INTRODUCTION
There is a lack of information on the dietary requirements of ducks grown under Australian
conditions with little having been done since the work of Siregar et al. (1982a, b). These
researchers reported that ducks needed a crude protein content of 18.7% in the starter diet (fed
days 1-14) and 16% (fed days 15-42) in the grower diet. Presently, diets fed to modern strains of
Pekin ducks contain more protein than these recommendations being around 22 and 19%,
respectively. Providing excess protein is wasteful and expensive, with the surplus protein
metabolised and used as a source of energy (Siregar, et al., 1982a). Evidence from a study in
2007, funded by RIRDC, indicated that ducks fed 19% protein in the grower diet had low
nitrogen retention, suggesting that this concentration was too high. In 2008, a further study
looked at the performance of commercial ducks fed a starter ration containing 18.1% and grower
ration with 15.7% protein. This was 3% less protein in both rations, than was being used by
industry at the time. In the first four weeks, ducks on the low-protein diets had significantly less
weight gain than ducks on the commercial control diet but remarkably, in week five, ducks on
the low-protein diet gained as much weight as ducks on the commercial control grower diet and
in week six gained more weight. These findings suggested that less protein than presently used in
commercial diets could be fed in weeks five and six of the production period and possibly
warrants the introduction of a finisher diet in commercial production.
1
Faculty of Veterinary Science, University of Sydney, Camden NSW, 2570.
186
Aust. Poult. Sci. Symp. 2010...21
Due to genetic selection occurring in different countries, there has been diversification of
the Pekin breed and development of specific commercial strains. Strains of particular
importance to Australian duck-meat production are the Cherry Valley (CV) and the Grimaud
Frère (GF). These strains have different growth characteristics but really neither meets the very
tight specification required by the Australian market which is for a 2.85-2.90 kg at six weeks of
age.
The objective of the present study was to investigate the dietary protein requirements of
three current duck strains, during weeks 5 and 6 of age, needed to obtain best performance and
growth under Australian summer conditions.
II. MATERIAL AND METHODS
One line of GF and two lines of CV Pekin duck strains were used in the study. For ‘commercialin-confidence’ reasons, the strains are identified as X, Y, and Z. The three lines were bred and
eggs incubated at a commercial breeder farm owned by our industry partners PEPE’S Duck’s Pty
Ltd. At hatch, ducklings were vent sexed and delivered to the University of Sydney’s Poultry
Research Unit in Camden. On arrival, eighteen males and eighteen females of the same strain
were randomly selected and weighed as a group prior to placement in their designated pen. Birds
were maintained in 48 floor pens (3m x 1.5 m), four replicate pens/treatment, in a tunnel
ventilated shed. Ducks were maintained in continuous lighting for the duration of the growth
trial. They were raised on deep litter, consisting of dry wood shavings. At placement, the
brooding temperature was 32°C and this was gradually reduced to 20°C at four weeks of age. On
day six, identification tags were inserted into the right wing of all ducks. During the six week
production period birds were removed at various times for growth and carcass measurements. At
the start of week six, 22 ducks remained in the pen (2045 cm2/bird).
Ducks had continuous access to both food and water. Diets were formulated and supplied
by Inghams Pty Ltd. A starter crumble containing 12.6 MJ/kg ME and 21.5% protein was fed to
all ducks from day 1 to day 14. The grower diet was fed as a pellet and contained 12.5 MJ/kg
ME and 17.5% protein and fed to all ducks from day 15 to day 28. On days 29 to 41, four
treatment finisher pelleted diets were fed which varied in protein content. Finisher diets
contained 12.5 MJ/kg ME with 16.8. 16.1, 15.4 and 17.5 % protein and were identified as diets
A, B, C and D, respectively. Diet D was considered the control as it had the same protein content
as the diet fed in the grower period.
Data were analysed using the REML linear mixed model function of Genstat® 11th
edition. Data was loge transformed where necessary. The fixed model included the effects of diet,
sex, strain and week and the interactions while the random model included the effects of block,
pen and bird identification. Significance testing of fixed effects was conducted using Wald chisquare tests with a significance threshold of P < 0.05. Any non-significant interactions were
removed from the model and for significant effects, the least significant difference (LSD) was
used to make pair-wise comparisons of means.
187
Aust. Poult. Sci. Symp. 2010...21
III. RESULTS
The effect of diet on liveweight gain (LG) and strain on liveweight (LW) are given in table 1.
There was a significant interaction between diet and week, (P < 0.001). No differences in LG
were seen in weeks 1 to 4. In week 5, LG was lower for birds fed diets B and C compared to
those fed diets A and D (P < 0.05). Interestingly, in week 6 these effects were reversed, with the
gain for birds fed diets B and C being greater than for birds fed diets A and D (P < 0.05). At the
end of the 6 week the only difference in LW was where birds on diet D were heavier than those
on diet C (P < 0.05).
Table 1.
The mean (± SEM) weekly LG and weekly LW for three strains (X, Y and Z) of
commercial Pekin ducks fed the same diets during the first 4 weeks and then diets
differing in protein content (diets A-D) during weeks five and six.
Diet
Week 1
A
216
±4
217
±4
214
±4
216
±4
B
C
D
Strain
Week 1
X
262
± 2b
270
± 2a
269
± 2a
Y
Z
Liveweight gain (LG)
(g/week)
Week 2
Week 3
Week 4
482
610
±9
± 11
483
627
±9
± 12
489
610
±9
± 12
478
614
±9
± 12
Liveweight (LW)
(g)
Week 2
Week 3
723
± 5b
756
± 6a
767
± 5a
1308
± 9b
1390
± 11 a
1390
± 10 a
Week 5
Week 6
611
± 12
626
± 12
609
± 11
626
± 12
587
± 11a
507
± 10b
525
± 10b
605
± 11a
407
± 8b
439
± 8a
440
± 8a
410
± 8b
Week 4
Week 5
Week 6
1878
± 13 b
2030
± 16 a
2026
± 14 a
2394
± 17 c
2617
± 21 a
2594
± 18 b
2799
± 20 c
3077
± 25 a
3022
± 21 b
There was a significant interaction between strain and week, (P < 0.001). Except for
week 1, the LW of strain X was less than for other strains (P < 0.05). For the first 4 weeks there
was no difference in LW of strains Y and Z. However, in weeks 5 and 6 the LW of strain Y was
greater than for strain Z (P < 0.05).
Diet had no effect (P = 0.18) on the feed conversion efficiency (FCR). For diets A, B, C
and D the FCR was 2.08 ± 0.02, 2.03 ± 0.02, 2.00 ± 0.02 and 2.00 ± 0.02, respectively. Strain
had a significant effect on FCR (P < 0.001) with strain X (2.06 ± 0.01) having lower efficiency
(P < 0.05) than strains Y (1.99 ± 0.01) and Z (1.99 ± 0.01).
188
Aust. Poult. Sci. Symp. 2010...21
IV. DISCUSSION
A major challenge for the duck-meat industry at the moment is formulating diets that meet duck
requirements for optimal growth in a cost-effective manner. While there were dietary effects
influencing weight gain during week five and six these were not consistent and tended to balance
out over the two week period such that the total weight gain over weeks five and six were similar
for all diets. This suggests that there is no benefit in feeding more than 16% protein in the diet
during weeks five and six. This would have economic benefits as feed makes up 70% of total
production costs in the duck-meat industry and protein is an expensive component of the feed
(Cherry and Morris, 2008). As there were no dietary effects on FCR, the data indicate that
producers can reduce costs by introducing a low protein finisher diet into the feeding program of
commercial ducks.
Previous studies in ducks have revealed that genotype has a significant effect on growth
rates (Maruyama, et al., 1999; Cherry and Morris, 2008). Ducks of strain X grew slower than the
other strains and were smaller at market age. In fact they failed to meet the market specification
of 2.9 kg and in commercial practice would need to be grown to an older age before processing.
Strain Y reached the heaviest weight at market age but the difference with strain Z was not great,
on average this being 55g. While it is not part of the results discussed here, strain Z had a
significantly higher breast muscle yield at six weeks than did strain Y (average: 359 Vs 286 g).
This is a major advantage for strain Z over strain Y and indicates that it is possible to select for
improved meat yield at an earlier age.
V. ACKNOWLEDGEMENTS
We acknowledge the financial support given by The Rural Industries Research and Development
Corporation (New Animal Industries) and the industry partners in this work, Pepe’s Ducks, Pty
Ltd. We also acknowledge the technical support given by staff at University of Sydney Poultry
Research Unit, Camden.
REFERENCES
Cherry, Morris TR (2008) Domestic Duck Production. CABI publishing, Wallingford.
Maruyama K, Akbar, MK, Turk CM (1999) British Poultry Science 40, 233-239.
Siregar AP, Cumming RB, Farrell DJ (1982a) Australian Journal of Agricultural Research 33,
857-864.
Siregar AP, Cumming RB, Farrell DJ (1982b) Australian Journal of Agricultural Research 33,
865-875.
189
Aust. Poult. Sci. Symp. 2010...21
TWO-DAY-OLD DUCKLINGS INTERACT MORE WITH A BELL DRINKER THAN A
NIPPLE DRINKER SUSPENDED ABOVE A TROUGH
G.M. CRONIN 1, K.J. WILLIAMS1 and J.A. DOWNING1
Farming ducks for meat production is increasing in Australia. In Europe, the welfare issues
associated with intensification of meat duck production were reviewed by Rodenburg et al.
(2005), who identified the manner in which water was provided was a potential welfare issue.
Specifically, concerns were raised whether ducks require access to ‘open water’ for their
welfare, since open water stimulated the performance of preening, dabbling, head-dipping,
bathing and swimming (Rodenburg et al., 2005). However, a consequence of water-related
behaviours was that more water may be used, resulting in increased spillage and reduced
litter quality. Cooper et al. (2002) investigated the behaviour of young ducks provided open
water via bell drinkers compared to nipple drinkers. Young ducks had a clear preference for
bell drinkers and placed a higher value on wider, deeper drinkers that allowed a greater range
of drinker-related activities than nipple drinkers alone. The objective of the present
experiment was to investigate the preference of 2-day-old ducks for two water presentation
systems, which provided different levels of open water but which, in principle, were
constructed using similar water-holding structures that permitted the ducks to sit in a trough.
Six pens of 36 ducklings (Cherry Valley and Grimaud Freres) were continuously video
recorded from the time of placement in pens at day-old. The ducklings were restricted to an
area of ~3.1 m2 within pens measuring 3.0 m x 1.5 m in an environment controlled shed.
Lighting was continuous and heating was provided in each pen by an infra-red globe heater
suspended 0.6 m above the floor, which was 50 mm deep wood shavings. Feed was available
ad libitum from a 40 cm diameter tray and a circular feeder. Water was provided by a bell
drinker positioned in the middle of the pen (Multiquip Pty Ltd, Austral, NSW, 13 l water
capacity, 35 cm diam) and three nipple drinkers with water catching ‘cups’ about 0.5 m apart,
suspended above a trough on one side of the pen. The number of ducklings interacting with
the bell drinker and the trough was collated from the digital video record at 5-min intervals
for 24 h commencing at 1200 h on the second day of life. Interaction with the water facility
was defined as ducklings having their head adjacent to (within 2 cm) or over the bell drinker
or trough. The number of ducklings sitting in the bell drinker or trough was also recorded.
The data were analysed using a two-sample T-test (paired) in Genstat (Release 11.1 (2008)
VSN International Ltd., UK) and the experimental unit was the pen of ducklings.
The likelihood that ducklings were observed at the bell drinker was twice that for the
nipple drinker/trough system (17.9 vs. 8.6% of observations, respectively; t = 2.50, 5 df, P =
0.027), supporting the findings of Cooper et al. (2002) that open water may be a visual
stimulus increasing ducks’ level of interaction with the water supply apparatus. However,
there was no difference in the incidence of ducklings sitting in the bell drinker (i.e. in the
water) compared to in the trough beneath the line of nipple drinkers where some spillage may
collect (1.6 vs. 1.5% of observations, respectively; t = 0.30, 5 df, P = 0.39). Further research
is required to investigate the relative stimulus values of the bell drinker and the trough, and
the relationship between the performance of water-related behaviours and welfare.
Cooper JJ, McAfee L, Skinn H (2002) Brit. Poult. Sci. 43 (suppl.) S17-S18.
Rodenburg TB, Bracke MBM, Berk J, Cooper J, Faure JM, Guémemé D, Guy G, Harlander
A, Jones T, Knierim U, Kuhnt K, Pingel H, Reiter K, Servière J, Ruis MAW (2005) World’s
Poult. Sci. J. 61, 633-646.
1
Faculty of Veterinary Science, The University of Sydney, Camden NSW 2570.
190
Aust. Poult. Sci. Symp. 2010...21
MICROBIAL PROFILES IN THE GASTRO-INTESTINAL TRACT OF BROILERS AND
ITS RELATION TO FEED EFFICIENCY
L.L.M. DE LANGE 1 and P.J.A. WIJTTEN 2
Summary
In a field survey and in a controlled experiment the gut microbiota of broilers was studied
using fragment lengths of DNA from genes encoding 16S rRNA to reveal the relation
between the composition of the microbiota in the gut and the performance of broilers. The
relation between feed utilisation and microbiota in the proximal part of the intestine of the
young broiler is more pronounced than on a later age in the more distal parts. The relative
quantities of DNA from L. acidophilus and probably L. salivarius in the crop and ileum,
expressed as percentage of the total microbial DNA, are negatively correlated with feed
efficiency.
I. INTRODUCTION
The ban on antibiotic growth promoters (AGP’s) since 2006 in the European Union has
increased the interest in gut microbiota composition of broilers and its effect on animal
performance. Many research projects have started to find alternatives for these AGP’s.
Knowledge about the effect of AGP’s on microbiota in the gastro-intestinal tract (GIT)
however is limited, and the relationship between the microbiota in the GIT and animal
performance has not been clarified very well.
Recently, new DNA-techniques to study the microbiota have been developed. In the
current study, the Terminal Restriction Fragment Length Polymorphism (T-RFLP) technique
described by Liu et al (1997) was used to study the microbiota in the GIT. The objective of
this research was to improve our understanding about the microbiota in the different
segments of the intestinal tract at different ages in broilers and how this relates to feed
utilisation.
The Dutch Ministry of Economic Affairs has supported this research financially.
II. MATERIALS AND METHODS
To reveal the microbiota in the different parts of the GIT of broilers two studies were
performed: one big field survey on 23 broiler farms with either a good or moderate
performance and one controlled experiment with different feeds on a test farm. On the 23
farms, samples were taken from the content of the crop, ileum and caeca at 7, 27 and 35 days
of age. Per farm, segment and age, 10 samples were taken. Thus, in total 2070 samples from
the field experiment were analysed.
In addition, in a controlled experiment with 1200 birds divided over 48 pens and
eight treatments with different starter and grower diets, samples were taken from 6 birds per
pen from the content of the crop and the ileum at day 9 and 29. Subsequently 1152 samples
from the controlled experiment were analysed on DNA fragment lengths.
1
De Heus Feeds BV, P.O. Box 396, 6710BJ Ede, The Netherlands
Provimi BV, P.O. Box 5063, 3008AB Rotterdam, The Netherlands
2
191
Aust. Poult. Sci. Symp. 2010...21
The 16S rRNA genes of the micro-organisms present in the sample were amplified
by PCR using blue fluorescently labelled primer 8f (5'-AGA GTT TGA TCC TGG CTC AG3') and yellow fluorescently labelled primer 926r (5'-ACC GCT TGT GCG GGC CC-3') and
were subsequently digested with restriction endonucleases MspI and HinpI (both from New
England Biolabs, Ipswich, MA, USA). This technique provides two outcomes per sample:
one from the blue and one from the yellow primer. The results were expressed as percentage
of the total DNA, per fragment length (FL). The FL varies from 50 to 600 nucleotides or base
pairs (BP). Databases were available to determine which lengths, depending on the used
primers and restriction enzymes, correspond with certain bacterial species. Because the data
from the blue primer were more discriminating for the bacterial species than the data from the
yellow primer, only the results of the blue primer are presented in this paper. For
identification of the bacterial species, the data from both primers were used and some
samples were sequenced with the kind help of M. Lee at the lab of Poultry Science from the
University of Georgia, USA.
III. RESULTS
Relative quantity DNA in ileum with 181 BP (%)
(a) Field survey
Only a few big peaks with 10–20% of the total DNA, most probably representing
Lactobacillus species, are observed in the microbial DNA from the crop and ileum. In the
caeca many small peaks are observed with no peak reaching more than 7% of the total DNA.
The resemblance in microbial composition between the crop and ileum is large, whereas the
composition in the caeca differs very much from that in the crop and ileum. The interaction
between the effect of farm type and age on the relative quantity of DNA with 181 BP is
significant (P < 0.05) as is shown in figure 1.
30
28
26
24
22
20
18
Farm
16
Good
14
Moderate
7
27
35
AGE (days)
Figure 1.
Relative quantity of DNA in the ileum with 181 BP on good and moderate
farms at different age
192
Aust. Poult. Sci. Symp. 2010...21
(b) Controlled experiment
In this experiment, the focus was on the relationship between feed efficiency (FE) and the
relative abundance of DNA with a certain FL. The relationship between FE in the starter
period and the relative DNA quantities of certain fragment lengths at day 9 were weak. The
relationship between FE and DNA abundance was more pronounced when the FCR during
the starter and grower period was related to the T-RFLP data on day 29. The relative quantity
of DNA with a FL of 188 BP in the crop had a positive relationship with FE, while fragments
with approximately 571 -573 BP in the crop showed a negative relationship with FE (Figure
2).
Relative quantity of DNA in the crop (%)
35
30
FL188; R2 = 0.90
25
20
15
FL 572; R2 = 0.95
10
5
0
0.670
0.675
0.680
0.685
0.690
0.695
0.700
Feed efficiency
Figure 2.
Relation between relative abundance of DNA with certain fragment lengths
with FE
(c) Identification
The peak with a FL of 181 BP with the blue primer corresponded well with the peak with a
FL of 345 BP with the yellow primer. Based on databases including the above-mentioned
primers and restriction enzymes, the sequencing of some samples and the literature review by
Lu et al. (2003), this peak most likely represents the species Lactobacillus acidophilus. It is
not clear which bacterial species represents the peak with a FL of 188 BP. It might be L.
crispatus, L. johnsonii, L. gasseri or a mixture of these species. Most probably, the peak with
circa 572 BP represents L. salivarius.
IV. DISCUSSION
One should realise that the used primers do not cover all bacterial species and that
identification with T-RFLP-technique with only two primers is not 100% reliable. Firstly, is
difficult to determine the exact length of especially long fragments. Secondly, more than one
species can have the same primed FL. Also the studied fragments can occur more than once
on the DNA of one bacterial species. Sequencing more samples will provide more reliable
information about identification. With good reviews about the main species that are common
193
Aust. Poult. Sci. Symp. 2010...21
in the GIT of broilers, the T-RFLP-technique can give a good view on the development of the
microbial population in the different segments of the gut of broilers.
From this research it cannot be concluded that the bacterial species L. acidophilus and
L. salivarius have caused the negative effect on the feed utilisation, although the relative
abundance of these species was clearly associated with FE. L. acidophilus is a bacterial
species, which can multiply very fast and might be an indicator for bacterial overgrowth. All
microbes compete with the host for nutrients, and also Lactobacilli trigger the immune
system with as possible consequence a decreased feed efficiency.
IV. CONCLUSIONS
The T-RFLP technique provides good information about the development of the microbial
composition in the gastro-intestinal tract during the lifetime of a broiler. The relation between
feed utilisation and microbiota in the proximal part of the intestine of the young broiler is
more pronounced than on a later age in the more distal parts. The relative quantities of DNA
from L. acidophilus and probably L. salivarius, expressed as percentage of the total microbial
DNA in the GIT, are negatively correlated with feed utilisation.
REFERENCES
Liu WT, Marsh TL, Cheng H and Forney LJ (1997) Applied and Environmental
Microbiology 63, 4516-4522.
Lu J, Idris U, Harmon B, Hofacre C, Maurer JJ, and Lee MD (2003) Applied and
Environmental Microbiology 69, 6816-6824.
194
Aust. Poult. Sci. Symp. 2010...21
CHARACTERISATION OF GUT BACTERIA ASSOCIATED WITH BROILER
PERFORMANCE ACROSS VARIOUS AUSTRALIAN FEEDING TRIALS
V.A. TOROK 1, K. OPHEL-KELLER1, L.L. MIKKELSEN 2, R. PEREZ-MALDONADO 3,
K. BALDING 4, R. MACALPINE4 and R. J. HUGHES 5.
Summary
Linkages were established between performance and gut microbiota from three poultry trials
investigating effect of various diets and/or litter on broiler performance. Bacterial profiling
was done to investigate changes in gut microbiota and identify potential performance related
bacteria. Across all three trials four operational taxonomic units (OTU) within the ileum
(180, 492, 564-566 and 936-938) and five OTU within the caeca (140-142, 216-222, 280286, 312 and 482) were consistently identified. Among these OTU 492, 140-142 and 482
were more closely associated with improved performance, while OTU 564-566 was mainly
associated with decreased performance. Targeted cloning and sequencing of 7 of these OTU
revealed a possible 19 different bacteria species, indicating many of these OTU contain
several bacteria which may be contributing to the increase or decrease of a particular
performance related OTU. Many of these bacteria were identifiable to the species level;
however, the majority remain unclassified bacteria. Where bacteria were identifiable to the
phyla level, they belonged predominantly to the Firmicutes and Bacteroidetes.
I. INTRODUCTION
A favourable gut microbiota is important for the optimal growth and performance of chickens
and several gut bacteria isolated from chicken have been shown to have various important
biochemical properties. Alternatively, an unfavourable microbiota may promote clinical and
sub-clinical enteric infections, leading to decreased growth rates and increased mortality.
However, gut bacteria need not be pathogenic to impact negatively on bird performance and
production. For example, the toxic metabolites generated by the bacterial β-glucuronidase in
poultry have been shown to hinder performance or reduce feed utilization (Jin et al., 2000),
while ammonia, produced by proteolytic bacteria such as Clostridium spp., Enterococcus spp.
and Bacteroides spp, have toxic effects on enterocytes (Rehman et al., 2007). Furthermore,
some members of the genera Clostridium, Lactobacillus, Fusobactrium, Bacteroides,
Bifidobacterium, Peptostreptococus, and Streptococcus produce bile salt hydrolase compounds
which lower the detergent properties of bile acids in the emulsification of fat leading to growth
depression in chickens (Knarreborg et al., 2002, Rehman et al., 2007).
Recently, a direct correlation has been demonstrated between overall changes in gut
microbiota associated with non-starch polysaccharide degrading enzyme supplementation and
improved bird performance (Torok et al., 2008). The aim of this study was to determine if
particular broiler gut bacteria could be consistently linked with improved or decreased
performance across three performance trials.
1
SARDI, Plant and Soil Health, GPO 397, Adelaide, SA 5001
School of Environmental and Rural Science, University of New England, Armidale, NSW
2351
3
Poultry Research Centre, DPI&F, Alexandra Hills, QLD, 4161
4
Inghams Enterprises Pty Limited, Leppington, NSW, 2179
5
SARDI, Pig and Poultry Production Institute, Roseworthy, SA 5371
2
195
Aust. Poult. Sci. Symp. 2010...21
II. MATERIALS AND METHODS
Linkages were established with three poultry performance trials between 2007-2008. These
trials evaluated influence of various dietary and/or litter materials on poultry performance as
measured by feed conversion ratio (FCR). Within each trial diets were formulated to have
similar energy and nutrient specification although composition of raw ingredients may have
varied between dietary treatments. Trial 1 was done at the Queensland Poultry Research
Centre, Alexandra Hills Qld in 2007. This trial (Evaluation of sorghum grains from Qld and
NSW for broiler growth performance in a semi-commercial environment) was led by Dr R
Perez-Maldonado and supported by RIRDC (project DAQ-326A). Trial 2 was done at
Inghams Enterprises Pty Ltd., Leppington NSW in March 2008. Trial 2 (The effect of litter
and dietary fibre on gut development, nutrient digestibility and gut microbiota) was led by Dr
L Mikkelson and supported by the Australian Poultry CRC (project 06-18). Trial 3 was done
at Inghams Enterprises Pty Ltd., Leppington NSW in November 2008. Trial 3 was led by Dr
R McAlpine and K Balding and was evaluating commercial broiler feeds produced in various
Inghams’ feed mills across Australia.
Where significant differences were detected in FCR (mean ± SE) the poultry gut
microbiota was investigated in an attempt to identify bacterial indicators potentially linked
with broiler performance. A segment (3 cm) of tissue and associated digesta was collected
from the mid-point of the ileum and one caeca (n=12/treatment on day 42 of age for trial 1
and n=24/treatment on days 35 and 42 for trials 2 and 3 respectively). All samples were
stored at -20°C and freeze dried prior to nucleic acid extraction. Total nucleic acid was
extracted using a modified SARDI proprietary method. Bacterial communities were analysed
by terminal-restriction fragment length polymorphism (Torok et al. 2008). Multivariate
statistical methods were used to identify significant differences in bacterial community
composition between treatments within each experiment and identify potential operational
taxonomic units (OTU) associated with performance differences. OTU can represent
particular bacterial species or taxonomically related groups of bacteria. Identity of OTU of
interest were determined using targeted cloning and sequencing (Widmer et al. 2006)
III. RESULTS
In trial 1, FCR (21-42 days) was significantly better for birds raised on either a commercial
sorghum diet (1.85±0.01) or commercial sorghum diet supplemented with phytase
(1.86±0.01), as compared with a control wheat based diet supplemented with xylanase
(1.99±0.02). Microbial profiling of the caecal samples showed that there were no significant
(P>0.05) differences between dietary treatments. However, within the ileum there were
significant (P<0.05) differences in bacterial communities of birds fed the wheat plus xylanase
diet when compared to birds fed either the sorghum commercial diet or sorghum commercial
diet supplemented with phytase. OTU 76, 180, 468, 492, 564, 936 were identified as
contributing to differences in ileal microbial community composition between birds raised on
the wheat control diet and birds raised on either of the sorghum diets. OTU 76, 180, 468, 492
and 936 were more abundant in the better performing sorghum diets, while OTU 564 was
more abundant in the poorer performing wheat control diet.
In trial 2, FCR (0-35 days) was significantly better for female broilers fed a low fibre
diet and reared on hardwood litter (1.51±0.02) as compared to those fed either a low fibre diet
and reared on paper litter (1.58±0.02) or a high fibre diet and reared on hardwood litter
(1.56±0.02). Microbial profiling of the ileal samples showed there were no significant
(P>0.05) differences between any of the dietary/litter treatments. However, for both male and
female birds, caecal microbial communities were significantly (P<0.05) different between
196
Aust. Poult. Sci. Symp. 2010...21
birds fed a low fibre diet and reared on paper litter versus birds fed a low fibre diet and reared
on wood litter. OTU 92-94, 142, 198, 206-208, 216-218, 222, 282, 284-286, 312, 482, 542
and 522 were identified as contributing to differences in caecal microbial community
composition between female birds in these two groups. OTU 92-94, 142, 198, 206-208, 482,
542 and 522 were more abundant in the group with improved performance as measured by
FCR, while OTU 222, 282, 284-286, and 312 were more abundant in the lower performing
group.
In trial 3, FCR (corrected to 2.6kg live weight) was significantly better for broilers fed
diets produced in mills in Western Australian (1.55±0.01) and Tasmanian (1.50±0.01) as
compared with diets produced in mills in Queensland (1.58±0.01) and New South Wales
(1.60±0.01). Microbial profiling of the ileal and caecal samples showed that there were
significant (P<0.05) differences between microbial communities of birds on the two better
performing diets versus birds raised on the two poorer performing diets. Within the ilea OTU
180, 188, 454, 492, 506, 566, 576, 872 and 938 were identified as contributing to differences
in microbial composition between improved and poorer performing birds as measured by
FCR. OTU 454, 492 and 506 were more abundant in the higher performing groups, while
OTU 180, 188, 566 and 938 were more abundant in the lower performing groups. Within the
caeca OTU 140-142, 212, 218-220 284-286, 312, 482, 488 and 536 were identified as
contributing to differences in microbial composition between improved and poorer
performing chickens. OTU 140-142, 218-220, 284-286, 312, 482, 488, and 536 were more
abundant in the better performing groups, while OTU 212 was more abundant in the lower
performing groups.
IV. DISCUSSION
Linkages were established with three poultry performance trials investigating effect of
various feed and/or litter on broiler performance. Bacterial profiling was done to investigate
changes in gut microbiota and identify potential bacterial indicators linked with performance.
Across all three trials four OTU within the ilea (180, 492, 564-566 and 936-938) and five
OTU within the caeca (140-142, 216-222, 280-286, 312 and 482) were consistently
identified. Targeted cloning and sequencing of 7 of these OTU revealed a possibility for 19
different bacterial species, indicating many of these OTU contain several bacteria which may
be contributing to the increase or decrease of a particular performance related OTU. Many of
these bacteria were identifiable to the species level however, the majority remain unclassified
bacteria. Where bacteria were identifiable to the phyla level they belong predominantly to the
Firmicutes and Bacteroidetes. The relative abundance of the Bacteroidetes and Firmicutes
have been shown to differ in genetically predisposed obese mice versus lean mice
(Turnbaugh et al., 2006) indicating particular bacterial groups have increased capacity for
energy harvest. Although many of these potential performance related bacteria are
unclassified they do show high sequence similarity (90-100%) with other unclassified
bacteria in public genome sequence databases. Many of these identified similar sequences
have been obtained from studies investigating the relationship between the gut microbiome
and host metabolic phenotype, innate immunity and gut microbiota, gut microbiota in various
host species, and several unpublished molecular poultry gut microbiota studies.
V. CONCLUSIONS
These results are promising in our quest to identify gut bacterial species linked to broiler
performance, however further well-planned work is required to validate findings. Specific
197
Aust. Poult. Sci. Symp. 2010...21
tests for performance related bacteria would allow us to better evaluate dietary treatments in
achieving an optimal gut microbiota for broiler performance and production.
ACKNOWLEDGEMENTS
We would like to acknowledge the financial support of the Australian Poultry CRC.
REFERENCES
Jin, L.Z., Ho, Y.W., Abdullah, N. and Jalaludin, S. (2000). Poultry Science, 79, 886-891.
Knarreborg, A., Engberg, R. M., Jensen, S. K. and Jensen, B. B. (2002). Applied and
Environmental Microbiology, 68, 6425-6428.
Rehman, H.U., Vahjen, W., Awad, W.A. and Zentek, J. (2007). Archives of Animal Nutrition,
61, 319-335.
Torok, V.A., Ophel-Keller, K, Loo, M., Hughes, R.J. (2008). Applied and Environmental
Microbiology, 74, 783-791.
Turnbaugh, P. J., Ley, R. E., Mahowald, M. A., Magrini, V., Mardis, E. R., and Gordon, J. I.
(2006). Nature, 444, 1027–1031.
Widmer, F., Hartmann, M., Frey, B. and Kolliker, R. (2006). Journal of Microbiological
Methods, 66, 512-520.
198
Aust. Poult. Sci. Symp. 2010...21
EFFECTS OF PHYTOGENICS AND ZINC BACITRACIN ON PERFORMANCE AND
INTESTINAL HEALTH STATUS OF BROILERS
R. M. GOUS 1, T. STEINER 2 and R. NICHOL 3
Summary
A 35-day trial was carried out to determine effects of phytogenics or zinc-bacitracin on
broiler performance and intestinal lesion scores. 2016 Cobb and Ross day-old broilers were
assigned to four treatments: (1) negative control, (2) positive control: zinc-bacitracin, (3) NC
+ phytogenics and (4) Treatment 3 plus coccidiosis vaccination (Paracox-5). Birds were fed
starter (day 0–7), grower (day 8–21) and finisher (day 22–35) diets based on corn and
soybean meal. Performance parameters did not differ (P > 0.05) between Cobb and Ross
broilers. After 35 days the lowest body weight recorded was for the control treatment (1435
g), whereas the highest body weights were recorded for birds in Treatments 3 and 4 (1508
and 1503 g, respectively), these being significantly (P < 0.05) heavier than birds in the
negative and positive control. Total feed consumption over the five-week period did not
differ (P > 0.05) between treatments. Feed conversion efficiency (FCE), however, was lowest
in Treatment 1 (468 g/kg), and differed numerically (P > 0.05) from Treatment 3 (516 g/kg).
In contrast, FCE was improved (P < 0.05) in comparison with the control in Treatments 2 and
4 (527 and 519 g/kg, respectively). Feed consumption prior to each feed change did not differ
between treatments (P > 0.05), but on the day of the first change (day 7) Treatment 1
recorded the lowest intake (29.5 g), whilst birds on Treatment 4 consumed the most feed
(32.2 g; P < 0.05). Feed intake in treatments 2 and 3 (31.1 and 31.5 g, respectively) were
higher (P < 0.05) than in the negative control, but numerically lower than in Treatment 4 (P >
0.05). There were no differences in feed intake at the second feed change (day 21).
Application of the coccidial vaccine in Treatment 4 did not affect performance parameters (P
> 0.05). In spite of small differences in airsac and trachea lesions, the health status of the
birds was excellent throughout the trial. In conclusion, phytogenics and zinc-bacitracin
beneficially affected performance of broilers reared under good hygienic conditions with
little intestinal challenge.
I. INTRODUCTION
Phytogenic compounds are regarded as potential modifiers of gut function and performance,
hence representing potential alternatives to antimicrobial growth promoters such as zincbacitracin. Indeed, a growing number of scientific reports show that these compounds exert
performance-enhancing effects in poultry (Mountzouris et al., 2009; Windisch et al., 2008).
The objective of this study was to investigate two areas related to gut function and
destabilization during periods of intestinal stress, and the potential of a phytogenic feed
additive or and zinc bacitracin to alleviate this stress. The first deals with stress related to the
change-over from one feed to the next, and the other, the effect of vaccinating for coccidiosis.
1
Department of Animal Science and Poultry Science, University of KwaZulu-Natal, Pietermaritzburg,
RSA
2
Biomin Holding GmbH, Herzogenburg, Austria
3
Biomin Singapore Pte Ltd, Singapore.
199
Aust. Poult. Sci. Symp. 2010...21
II. MATERIALS AND METHODS
2016 as-hatched Cobb and Ross broilers, treated at day old with IB-H120 and ND-VH, were
used in the trial. They were placed 63 to each of 48 floor pens in a tunnel ventilated broiler
house at an initial stocking density of 14 birds/pen. Cross ventilation was used initially to
provide fresh air, and gas brooders, centred over the junction between four pens, were used to
provide warmth. Control over these conditions was automatic according to the settings
applied. After 14 days, use was made of longitudinal ventilation when necessary.
A commercial feed (Meadow Feeds, Pietermaritzburg) without antibiotics, growth
promoters or coccidiostats, was used as control. Three phases were implemented: Starter,
Grower and Finisher and the trial was terminated at 35 d. The starter diet, in mash form, was
fed for 7 d. Feed remaining in the troughs at that point was removed and the grower diet
introduced from 8–21 d. Grower and Finisher diets were pelleted. The following treatments
were implemented (1) Negative control (NC) diets, (2) Positive control + AGP (Zinc
Bacitracin at 125 g/t), (3) NC + Phytogenics (Biomin® P.E.P. 125 poultry, 125 g/t) and (4)
NC + Phytogenics + coccidiosis vaccination. In Treatment 4 ParacoxTM- 5 was used to
vaccinate the broilers against coccidiosis. Access to nipple drinkers in pens receiving
coccidiosis vaccine was barred. Water with vaccine was given in open troughs. None of the
birds was treated with a coccidiostat during the trial, nor was it necessary to treat them
prophylactically.
Body weight and food intake were recorded at 7, 21 and 35 d. Vent scoring to indicate
presence of loose droppings was done on a daily basis. Daily measurements of feed intake
were carried out for one day before and three days after the two periods of feed change by
weighing the feeders on days 6–9 and 20–23. Lesion score of the intestines was conducted on
ten birds per treatment at the end of the trial. The measurements were made using a scoring
system that is used in practice for evaluating the efficacy of coccidiostats. Each bird was
scored for lesions in the trachea, airsacs and gizzard (erosion), for the presence of Eimeria
species, and for the status of the Bursa of Fabricius.
III. RESULTS
Performance parameters did not differ (P > 0.05) between Cobb and Ross broilers. After 35
days the lowest body weight recorded was for the control treatment (1435 g), whereas the
highest body weights were recorded for birds in Treatments 3 and 4 (1508 and 1503 g,
respectively), these being significantly (P < 0.05) heavier than birds in the negative and
positive control (Figure 1). Total feed consumption was 3080, 2782, 2836 and 2812 g in
Treatments 1, 2, 3 and 4, respectively, and did not differ (P > 0.05) between treatments. Feed
conversion efficiency (FCE), however, was lowest in Treatment 1 (468 g/kg), and differed
numerically (P > 0.05) from Treatment 3 (516 g/kg) (Figure 2). In contrast, FCE was
improved (P < 0.05) in comparison with the control in Treatments 2 and 4 (527 and 519 g/kg,
respectively).
Feed consumption prior to the first feed change did not differ between treatments
(33.3–35.7 g; P > 0.05), but on the day of the first change (day 7) Treatment 1 recorded the
lowest intake (29.5 g), while birds on Treatment 4 consumed the most feed (32.2 g; P < 0.05)
(Figure 3). Feed intake in treatments 2 and 3 (31.1 and 31.5 g, respectively) were higher (P <
0.05) than in the negative control, but numerically lower than in Treatment 4 (P > 0.05).
There were no differences in feed intake at the second feed change (day 21). Application of
the coccidial vaccine in Treatment 4 did not affect performance parameters (P > 0.05).
The health status of broilers at the end of the trial was considered to be excellent
taking into account the relative absence of lesions in the airsacs, trachea and gizzard, and the
200
Aust. Poult. Sci. Symp. 2010...21
presence of only E. maxima in very few of the birds sampled (Table 1). The size of the Bursa
was consistent with healthy, uninfected birds in all cases. No statistical analysis was possible
on these data.
Figure 1.
Body weight as affected by antibiotics, phytogenics and phytogenics plus
coccidiosis vaccination. a,bMeans with a different superscript differ
significantly (P < 0.05).
Figure 2.
Feed Conversion Efficiency (FCE) as affected by antibiotics, phytogenics and
phytogenics plus coccidiosis vaccination. a,bMeans with a different superscript
differ significantly (P < 0.05).
Figure 3.
Feed intake per bird prior to and following the first change in feed
composition
201
Aust. Poult. Sci. Symp. 2010...21
Table 1.
Incidence of various lesions, and bursa size, of 10 birds from each treatment
Treatment
Airsac
Trachea
Gizzard
lesions1
lesions1
erosion1
1
2
0
1
2
2
0
1
3
5
1
1
4
3
0
0
1
Scoring: 0–5 (0 = no lesions × number of birds showing lesions)
2
Scoring: 0–6 (6 being a healthy Bursa for that age)
E. maxima
8
1
0
1
Bursa
Size2
5
6
6
5
IV. DISCUSSION
Changes in feed composition represent a potential stressor, which may result in a
destabilization of the gut due to a change in intestinal microflora. This destabilization often
presents as visual wet droppings for a few days and a lag in daily weight gain. This would be
demonstrated by a reduction in feed intake resulting from the feed change, which confirms
previous reports (Windisch et al., 2008). The addition of the phytogenic feed additive
improved performance of broilers to 35 d in this trial. Although the total amount of feed
consumed per bird did not differ between treatments, the improved growth rate at a similar
feed intake resulted in improved feed conversion efficiencies when this feed additive was
included. Effects were, in most cases, similar to those obtained with zinc-bacitracin. In
conclusion, phytogenics and zinc-bacitracin beneficially affected performance of broilers
reared under good hygienic conditions with little intestinal challenge. The additional feed
consumption in these treatments at the first feed change might have been partly responsible
for the heavier body weights of birds at the end of the growing period.
REFERENCES
Mountzouris KC, Paraskevas V, Fegeros K (2009) In: Phytogenics in Animal Nutrition. [T.
Steiner, ed.] pp. 97-110. Nottingham University Press.
Windisch W, Schedle K, Plitzner C, Kroismayr A (2008) Journal of Animal Science 86, 140148.
202
Aust. Poult. Sci. Symp. 2010...21
THE ABILITY OF GREEN TEA TO POSITIVELY MODIFY THE GUT
MICROFLORA IN BROILER CHICKENS
D.V. THOMAS 1, A.L. MOLAN1 and V. RAVINDRAN1
Summary
The objective of the current study was to investigate the influence of two green teas [normal
(N-GTE) and selenium-containing (Se-GTE) green teas] on changes in gut bacterial
population of broiler chickens. The effects of green teas on faecal and caecal microflora, and
faecal bacterial metabolic activities were studied in broilers fed a basal diet or basal diet
supplemented with 1.0% green tea for 35 days. Gut microbial communities were analysed by
the Florescent in Situ Hybridisation (FISH) technique at 7, 21, and 35 days of age. Dietary
inclusion of green tea had a positive effect on both feed efficiency and on the gut microflora
with an increased numbers of beneficial bacteria (Lactobacillus spp. and Bifidobacterium
spp.) and a reduced number of pathogenic bacteria (Clostridium spp. and Bacteroides spp.).
Although both teas showed beneficial effects on feed efficiency, Se-GTE was more beneficial
than N-GTE in the positive modulation of gut microflora. Relative to the control group, the
level of faecal bacterial ß-glucuronidase, an enzyme generated mainly by Escherichia coli,
Bacteroides spp., and Clostrdium spp., was decreased significantly while the level of ßglucosidase, an enzyme generated mainly by lactobacilli and bifidobacteria, was increased
significantly in the faecal contents of the chickens treated with both teas. Moreover, the pH of
the faecal contents collected from the birds fed diets with green teas was significantly lower
than that of the control group. In conclusion, the results of this study are beneficial in
developing and evaluating natural alternatives to in-feed antibiotics for sustainable poultry
production and also provide evidence that gut microbial community composition is
associated with growth performance.
I. INTRODUCTION
The role of commensal gut microflora in animal production is receiving much interest,
particularly with the global trend of moving away from in-feed antibiotics. The gut contains a
complex microbial community including potentially beneficial and pathogenic bacteria. An
unfavourable microflora may promote enteric infections, leading to decreased growth rates
and increased mortality. Consequently, a favourable gut microflora is necessary for the
optimal growth and performance of chickens.
Green tea is derived from the leaves and fine stems of the leguminous shrub, Camellia
sinensis with health promoting effects mainly attributed to its polyphenol content (Wolfram
et al., 2006) and antioxidant activity (Molan et al., 2009). Green tea has been shown in rodent
studies (Molan et al., 2007) to modify gut microflora, which may impact upon the
digestibility of nutrients, and affect chick performance. In this study changes in growth and
feed efficiency were measured and gut microbial communities analysed to assess the effect of
green tea dietary supplementation.
1
Institute of Food, Nutrition and Human Health, Massey University, Palmerston North
203
Aust. Poult. Sci. Symp. 2010...21
II. MATERIALS AND METHODS
Two experimental diets were prepared by the addition of one of two types of powdered green
tea to a wheat based control diet at an inclusion rate of 1% giving a total of three diets. The
green tea types used were either locally sourced Chinese green tea (N-GTE) or a high
selenium green tea (Se-GTE) sourced from China. The diets were formulated to meet or
exceed the NRC (1994) requirements for all nutrients for broilers and were pelleted at 70 0C.
Each diet was fed to six cages (8 birds/ cage). The birds received fluorescent illumination for
20 hours per day, and were allowed free access to the diets and water. Body weights and feed
intakes were measured weekly. In addition, a cohort study (n=192) was undertaken to provide
sampling of digesta on days 7, 21 and 35.
Bacterial measurement was undertaken on days 7, 21 and 35 using fluorescent in situ
hybridization (FISH) analysis of microflora. The probes used in the study were specific for
Lactobacillus spp., Bifidobacterium spp., Clostridium spp., and Bacteroides spp. These were
commercially synthesised and labelled with the fluorescent dye Cy3 (GeneWorks, Australia).
The procedure described by Dinoto et al. (2006) was followed with some modifications. The
β-glucuronidase and β-glucosidase activities were determined aerobically according to the
method of Goldin et al. (1980) and as described by Preter et al. (2008) with some
modifications. Faecal pH was measured on days 7, 21 and 35. The pH was measured using a
digital pH-meter at room temperature (20-22 oC).
Populations for each bacterial group were expressed as log number of bacterial cells/
gram caecal materials. Logarithmically-transformed data were analysed by one way analysis
of variance using SAS (version 9.1) with the level of significance set at
P < 0.05.
III.RESULTS AND DISCUSSION
Normal Chinese green tea (N-GTE) supplementation reduced (P < 0.05) weight gain during
the 5-week trial period, and high selenium green tea (Se-GTE) supplementation reduced (P <
0.05) weight gain on days 7 and 35. Feed intake was reduced (P < 0.05) in both treatment
groups over the whole trial period, but feed efficiency was improved (P < 0.05) as green tea
supplementation reduced feed intake to a greater extent than weight gain.
Dietary inclusion of green tea had a positive effect on the gut microflora (Table 1)
with an increased numbers of beneficial bacteria (Lactobacillus spp. and Bifidobacterium
spp.) and a reduced number of pathogenic bacteria (Clostridium spp and Bacteroides spp.).
Differences between the two green tea types were observed, with Se-GTE more beneficial
than N-GTE in the positive modulation of gut microflora. A number of studies have shown
that increasing the numbers of lactic acid bacteria (lactobacilli and bifidobacteria) in the
colon reduces the formation of ammonia, skatole, harmful amines and other procarcinogens
in the large intestine and the carcinogenic load on the intestine of humans (Burns and
Rowland 2000; Yamamoto et al., 1997).
204
Aust. Poult. Sci. Symp. 2010...21
Table 1.
Influence of dietary treatments on the enumeration of Bifidobacterium species,
Lactobacillus species, Bacteroides group and Clostridium species [Log10
cells/g of wet caecal contents] in caecal samples of broilers
Day 7
Day 21
Day 35
Log10
SE
Log10
SE
Log10
SE
Control
7.12
0.021
6.98
0.040
7.19
0.023
N-GTE
7.13
0.025
7.30***
0.044
7.27**
0.024
Se-GTE
7.20*
0.021
7.29***
0.029
7.32***
0.012
Control
6.06
0.029
5.91
0.062
5.68
0.059
N-GTE
6.06
0.039
5.98
0.045
6.16***
0.022
Se-GTE
6.16*
0.032
6.16**
0.040
6.20***
0.024
Control
7.14
0.032
7.00
0.027
7.08
0.020
N-GTE
6.96***
0.037
6.77***
0.036
6.98
0.058
SE-GTE
6.82***
0.039
6.77***
0.031
6.42***
0.043
Control
7.20
0.033
7.06
0.031
7.34
0.027
N-GTE
7.10
0.065
7.03
0.047
6.53***
0.034
Se-GTE
6.82***
0.070
6.98
0.030
6.22***
0.045
Bifidobacteria
Lactobacilli
Clostridia
Bacteriodes
* P < 0.05; ** P < 0.01 and *** P < 0.001 (significantly different from control).
Inclusion of both green teas in the diet resulted in a significant reduction (P < 0.01 to
0.0001) in the numbers of pathogenic bacteria. Physiological concentrations of green tea
polyphenols and extracts have been shown in in vitro studies to delay or inhibit the growth of
a wide range of pathogenic strains of enteric bacteria, including pathogenic strains of
Escherichia coli (Yam et al., 1997; Ciraj et al., 2001; Ishihara et al., 2001).
Relative to the control group, the level of faecal bacterial ß-glucuronidase, an enzyme
generated mainly by Escherichia coli, Bacteroides spp., and Clostridium spp., was decreased
significantly (P < 0.05) in the faecal contents of the chickens fed diets with both green teas.
Moreover, the level of ß-glucosidase, an enzyme generated mainly by lactobacilli and
bifidobacteria, was increased significantly (P < 0.05) in the faecal contents of the chickens
fed diets containing with both green teas when compared with the control group.
In addition, the pH of the faecal contents from birds fed diets with Se-GTE was
significantly lower (P < 0.05) than that of the control group. Supplementation with N-GTE
resulted in a significant decline (P < 0.05) in the pH at 21 and 35 days of age. The pH was
significantly lower in the faecal samples collected from birds supplemented with Se-GTE
than those supplemented with N-GTE after 7 days (P < 0.01) and 35 days (P < 0.05). The
lower pH in the faecal samples of birds fed diets with green teas, compared to the controls,
may be related to the growth of lactic acid bacteria.
205
Aust. Poult. Sci. Symp. 2010...21
IV. CONCLUSIONS
In conclusion, the present data suggest that green tea supplementation may be beneficial in
maintaining a healthy gut microflora in the absence of in feed antibiotics and assist in
efficient feed utilisation in broilers.
REFERENCES
Burns AJ, Rowland IR (2000) Current Issues in Intestinal Microbiology 1, 13-24.
Ciraj AM, Sulaim J, Mamatha B, Gopalkrishna BK, Shivananda PG (2001) Indian Journal of
Medical Science 55: 376–381.
Dinoto A, Suksomcheep A, Ishizuka S, Kimura H, Hanada S, Kamagata Y, Asano K, Tomita
F, Yokota A (2006) Applied and Environmental Microbiology 72, 784-792.
Goldin BR, Swenson L, Dwyer J, Sexton M, Gorbach SL (1980) Journal of the National
Cancer Institute 64, 255–261.
Ishihara N, Chu D, Akachi S, Juneja LR (2001) Livestock Production Science 68, 217–229.
Molan AL, De S, Meagher L (2009) International Journal of Food Sciences and Nutrition
(DOI: 10.1080/09637480701781490).
Molan AL, Flanagan J, Moughan P (2007) Proceedings, 3rd International Conference on ocha (tea) Culture and Science. Japan.
Wolfram S, Wang Y, Thieleke F (2006) Molecular Nutrition and Food Research 50, 176187.
Yam TS, Shah S, Hamilton-Miller JM (1997) FEMS Microbiology Letters 152, 169–174.
Yamamoto T, Juneja LR, Chu DC, Kim M (1997) Chemistry and Applications of Green Tea.
CRC press LLC, Boca Raton, USA.
206
Aust. Poult. Sci. Symp. 2010...21
EFFECTS OF NOVEL FEED ADDITIVES ON GUT HEALTH AND OVERALL
PERFORMANCE IN BIRDS CHALLENGED WITH CLOSTRIDIUM PERFRINGENS
M.S. GEIER 1, L.L. MIKKELSEN 2, V.A. TOROK 3, G.E. ALLISON 4, C. OLNOOD2,
A. SETIA2, M. CHOCT 5, M. BOULIANNE 6 and R.J. HUGHES1
Summary
The capacity for Lactobacillus johnsonii and an organic acid (OA) blend to prevent necrotic
enteritis (NE) was studied. Additionally, we evaluated the influence of Clostridium
perfringens challenge, zinc bacitracin (ZnB), L. johnsonii and OA on the intestinal
microbiota. Cobb 500 birds were allocated into six groups; unchallenged (Control),
challenged (Cp), zinc bacitracin (ZnB), organic acid (OA), L. johnsonii, and vehicle (n = 25
birds/pen, 8 pens/treatment). All birds were challenged with C. perfringens except for the
Control group. Only birds fed ZnB were protected from NE as indicated by maintenance of
body weight, low mortality and clostridia levels, and decreased intestinal macroscopic lesion
score compared to Cp-challenged controls. L. johnsonii-fed birds had reduced lesion scores
whilst OA-fed birds had reduced clostridia levels. Both L. johnsonii and OA-fed birds had
improved feed conversion ratios; however, mortality and body weights were not improved by
either treatment. Microbial profiling indicated that C. perfringens significantly altered the
jejunal microbiota. The microbiota of ZnB-fed birds was different to all other treatments.
Whilst OA and L. johnsonii altered some intestinal parameters, no protection against NE was
observed. The search for alternatives to antibiotics is important for the poultry industry.
Knowledge of the mechanisms involved in ZnB-mediated protection may lead to the
identification of compounds (or combinations) that promote a similar intestinal environment.
I. INTRODUCTION
Necrotic enteritis (NE) caused by the bacterium Clostridium perfringens, is one of the
world’s most prominent and severe poultry diseases. The economic impact of NE on the
world-wide poultry industry is estimated at over $2 billion per annum (Choct and Kocher,
2008). NE is typified by intestinal lesions, diarrhoea, impaired digestive function, reduced
nutrient absorption, and decreased feed intake (Kocher et al., 2004). In its acute form, NE can
cause significant flock mortality over a period of several days whilst sub-clinical NE can
significantly impair bird performance.
Currently in Australia, NE is controlled by the in-feed supplementation of antibiotics
such as zinc bacitracin (ZnB). The European Union have enforced a ban on the use of in-feed
antibiotics, and consumer pressure in other regions may force similar restrictions on
antibiotic use. Therefore, alternative strategies for the control of NE are needed which can
limit the health and economic impacts of the disease. The aims of this study were to assess
the capacity for an organic acid (OA) product and a candidate probiotic organism,
Lactobacillus johnsonii, to reduce the severity of NE in broilers. In addition, we aimed to
assess the effects of these compounds on the intestinal microbiota.
1
SARDI, Pig and Poultry Production Institute, Roseworthy, SA 5371
School of Environmental and Rural Science, UNE, Armidale, NSW 2351
3
SARDI, Plant and Soil Health, Adelaide, SA 5001
4
School of Biochemistry & Molecular Biology and ANU Medical School ANU Canberra ACT 0200
5
Australian Poultry CRC, University of New England, Armidale, NSW 2351
6
Faculty of Veterinary Medicine, University of Montreal, Québec, Canada
2
207
Aust. Poult. Sci. Symp. 2010...21
II. MATERIALS AND METHODS
Twelve hundred male Cobb 500 birds were randomly allocated to 48 pens (n=25 birds/pen)
and assigned to six treatment groups (8 pen replicates/treatment); unchallenged (Control),
Clostridium perfringens-challenged (Cp), ZnB (45 ppm + 100 ppm monensin in the diet), OA
(2 kg/ton in the diet), L. johnsonii (109 cfu/ml in PBS, administered orally on days 1, 3, 7, and
12), or vehicle control (PBS gavage on corresponding days). All groups were challenged with
C. perfringens except for the Control group. The NE challenge procedure was carried out as
described previously (Kocher et al., 2004; Mikkelsen et al., 2009). Birds were fed a starter
diet from placement until day 7, and again between days 15-22, with finisher diet from day
23 until completion of the trial. Between days 7-15, birds were fed a high protein diet (50%
fishmeal and 50% starter) to facilitate NE development. On day 9, all birds (except
unchallenged controls) were given a suspension of 2,500 oocysts of Eimeria acervulina, E.
maxima and E. tenella in 1 ml PBS. On day 15, birds in challenged groups were individually
inoculated with 1 ml of C. perfringens at a concentration of 3.5 x 108 cfu/ml.
Live weight and feed consumption were recorded on days 0, 9, 14, 21 and 28 and feed
conversion ratios were calculated. On days 15 and 18 two chickens per pen (n=16
birds/treatment) were selected for intestinal lesion scoring and clostridia enumeration. The
small intestine from each bird was incised longitudinally and examined for gross necrotic
lesions. Lesions were scored according to the previously described criteria (Prescott et al.,
1978). Clostridia enumeration was performed on days 15 and 18 as described previously
(Mikkelsen et al., 2009). On day 18, 2 birds in each pen (n=16 birds/treatment) were killed
for tissue collection. A segment (3 cm) of tissue and associated digesta was collected from
the mid-point of the jejunum and stored at 4°C until later frozen for microbial profiling by
terminal-restriction fragment length polymorphism (T-RFLP) analysis and assessment of
Lactobacillus species by denaturing gradient gel electrophoresis (Lac PCR-DGGE).
III. RESULTS
On day 9, prior to C. perfringens challenge, birds treated with L. johnsonii had a significantly
greater body weight compared to birds fed basal diet (Unchallenged and Cp-challenged
control groups; P<0.05; Table 1). On days 21 and 28, birds fed ZnB had the greatest body
weights compared to all other Cp-challenged groups (P<0.05). The mean body weight of
ZnB-fed birds was comparable to unchallenged birds (P>0.05). Birds fed ZnB had a lower
FCR compared to other Cp-challenged groups in the periods between days 0-21 and 0-28
(P<0.05). Birds fed OA had a significantly lower FCR compared to Cp-challenged controls
between days 0-21 and 0-28 (P<0.05). In the period between days 0-28, birds treated with L.
johnsonii also had a lower FCR compared to Cp-challenged controls (P<0.05).
In the period following C. perfringens challenge (days 14-21), bird mortality was
significantly elevated in Cp-challenged controls, vehicle controls, OA and L. johnsoniitreated groups compared to unchallenged birds and challenged birds fed ZnB (P<0.05; Table
1). On day 18, ZnB and OA treatment significantly reduced clostridia levels compared to Cpchallenged controls (P<0.05; Table 2). On day 18, duodenum macroscopic lesion scores were
significantly greater in Cp-challenged controls, vehicle controls and OA-fed birds compared
to unchallenged controls (P<0.05). ZnB and L. johnsonii-treated birds had significantly lower
duodenum lesion scores compared to Cp-challenged controls, vehicle controls and OA-fed
birds whilst having a lesion score comparable to unchallenged birds (P>0.05). In the jejunum,
day 18 lesion scores were significantly greater in Cp-challenged controls, vehicle controls
and OA-treated birds compared to unchallenged controls (P<0.05). ZnB and L. johnsoniitreated birds had a significantly lower score compared to Cp-challenged controls (P<0.05).
208
Aust. Poult. Sci. Symp. 2010...21
Table 1.
Body weight, feed conversion and mortality
Control
Cp
ZnB
OA
Vehicle
Weight
0
37 ± 0.2
38 ± 0.4
38 ± 0.3
37 ± 0.2
37 ± 0.1
bc
c
abc
ab
9
196 ± 1
192 ± 5
201 ± 4
205 ± 2
201 ± 1abc
14
380 ± 4
368 ± 6
386 ± 5
380 ± 4
378 ± 5
a
b
a
b
21
800 ± 10
552 ± 26
817 ± 12
567 ± 20
580 ± 14b
28
1374 ± 17a 1008 ± 46b 1420 ± 14a 1048 ± 30b 1082 ± 15b
Mortality
0-9
0.5
3.0
3.0
1.5
2.0
9-14
2.5
1.0
0.5
3.0
1.0
a
b
a
b
14-21
5.5
36.0
7.5
30.5
35.5b
21-28
1.0
0.0
2.5
1.0
0.0
FCR
0-21
1.29 ±
1.58 ±
1.25 ±
1.47 ±
1.56 ±
c
a
c
b
0.01
0.06
0.01
0.02
0.04ab
0-28
1.41 ±
1.66 ±
1.37 ±
1.51 ±
1.59 ±
cd
a
d
bc
0.02
0.08
0.01
0.01
0.04ab
Lj
P
38 ± 0.3
208 ± 2a
383 ± 5
574 ± 13b
1065 ± 21b
**
***
***
2.0
1.5
35.5b
1.0
***
***
1.53 ±
0.03ab
1.51 ±
0.03bc
***
Live weight data (g) and feed conversion ratio (FCR) are expressed as mean ± SEM (n=8
pens/treatment). Mortality data are expressed as percentage (%). Values within a row that do not share
a common letter are significantly different (P<0.05). ** P<0.01; *** P<0.001. Cp, Clostridium
perfringens; Lj, Lactobacillus johnsonii; OA, organic acid; ZnB, zinc bacitracin.
Table 2.
Macroscopic intestinal lesion scores and clostridia enumeration
Duodenum lesions
Day 15
Day 18
Jejunum lesions
Day 15
Day 18
Clostridia (log10 cfu/g)
Day 15
Day 18
Control
Cp
ZnB
OA
Vehicle
Lj
P
0.0 ±
0.0a
0.0 ±
0.0a
0.84 ±
0.20bc
0.56 ±
0.17b
0.0 ±
0.0a
0.06 ±
0.04a
1.13 ±
0.17c
0.28 ±
0.10b
1.13 ±
0.15c
0.53 ±
0.17b
0.38 ±
0.17b
0.09 ±
0.09a
***
0.0 ±
0.0a
0.0 ±
0.0a
0.31 ±
0.14b
0.75 ±
0.27c
0.0 ±
0.0a
0.0 ±
0.0a
0.78 ±
0.27b
0.50 ±
0.18bc
0.59 ±
0.20b
0.38 ±
0.10bc
0.28 ±
0.13b
0.19 ±
0.11ab
**
3.50 ±
0.12
3.69 ±
0.28ab
3.81 ±
0.28
4.23 ±
0.35a
3.60 ±
0.20
3.45 ±
0.32b
5.14 ±
0.69
3.42 ±
0.16b
3.71 ±
0.43
4.16 ±
0.22a
3.47 ±
0.24
4.44 ±
0.35a
***
***
-
*
Lesion scores and clostridia levels (log10 cfu/g digesta) are expressed as mean ± SEM. Values within a
row that do not share a common letter are significantly different (P<0.05). * P<0.05; ** P<0.01; ***
P<0.001.
209
Aust. Poult. Sci. Symp. 2010...21
Unchallenged control birds had a significantly different microbial profile in the
jejunum compared to Cp-challenged controls (P<0.05; Table 3), indicating that the challenge
procedure had a direct impact on the overall microbiota. Unchallenged birds also displayed a
significantly different microbial profile compared to ZnB, OA and vehicle-treated birds
(P<0.05), with a strong trend toward a difference compared to L. johnsonii-treated birds
(P<0.1). The microbial communities of Cp-challenged controls were significantly different
compared to ZnB and vehicle-treated birds (P<0.05); however, they were not significantly
different compared to OA and L. johnsonii-treated birds (P>0.05). Birds fed ZnB had a
significantly different microbial composition compared to all other treatments (P<0.001).
Birds treated with OA, vehicle and L. johnsonii were not significantly different to one
another; however there appeared to be a trend toward a difference between OA and vehicletreated birds (P<0.1). C. perfringens challenge significantly influenced the Lactobacillus
communities of broilers (P<0.05; data not shown). Birds fed ZnB had a significantly different
Lactobacillus profile compared to all other treatments (P<0.05). Interestingly, L. johnsonii
was only detected in half of the birds treated with L. johnsonii on days 1, 3, 7 and 12.
Table 3.
Control
Cp
ZnB
OA
Vehicle
Lj
One-way analysis of similarities (ANOSIM) of microbial communities
Control
0.012
<0.001
0.014
0.021
0.078
Cp
0.134
<0.001
0.388
0.024
0.543
ZnB
0.404
0.299
<0.001
<0.001
<0.001
OA
0.132
0.004
0.457
0.084
0.483
Vehicle
0.127
0.103
0.546
0.060
0.513
Lj
0.066
-0.01
0.374
-0.005
-0.008
-
Data are expressed as R-statistic (bold), with significance level (italics). P<0.05 was considered
significant. The global R-value was 0.178 at a significance level of 0.0001 which is considered
significant.
IV. CONCLUSION
In-feed supplementation with L. johnsonii or OA did not successfully prevent the onset of NE
in broilers, whilst conventional treatment with ZnB prevented the development of NE and
maintained bird performance during the challenge. C. perfringens challenge appeared to
significantly alter the microbiota, and ZnB shifted the intestinal microbial communities
compared to all other treatments. The unique microbial profile produced by ZnB may be
associated with improved health and performance. The potential role of probiotics and OAs
in the prevention of NE remains unclear; however, the observations that OA decreased
intestinal clostridial load and L. johnsonii decreased lesion scores whilst both partially
improved FCR indicate that other compounds and organisms of this nature, and combinations
of OAs and probiotics may indeed have the capacity to reduce the severity of NE.
REFERENCES
Choct M, Kocher A (2008) World's Poultry Science Journal 64 (Supp 2), 159.
Kocher A, Choct M, Teo A, Tan HM, Carter RR (2004) Proceedings, Australian Poultry
Science Symposium 16, 84-87.
Mikkelsen LL, Vidanarachchi JK, Olnood CG, Bao YM, Selle PH, Choct M (2009) British
Poultry Science 50, 66-75.
Prescott JF, Sivendra R, Barnum, DA (1978) Canadian Veterinary Journal 19, 181-183.
210
Aust. Poult. Sci. Symp. 2010...21
FEED ADDITIVES INFLUENCE GOBLET CELL DISTRIBUTION AND VILLUSCRYPT ARCHITECTURE IN BROILERS AFTER NECROTIC ENTERITIS
CHALLENGE
H.M. GOLDER 1, M.S. GEIER 2, P.I. HYND1, R.E.A. FORDER1, M. BOULIANNE 3
and R.J. HUGHES2
Summary
The probiotic, Lactobacillus johnsonii and a commercial organic acid (OA) blend were
histologically evaluated as potential non-antibiotic preventatives for necrotic enteritis (NE).
A total of 1200 Cobb 500 broilers were randomly assigned to the following six treatment
groups; unchallenged (Control), Clostridium perfringens challenged (Cp), zinc bacitracin
(ZnB), organic acid (OA), vehicle and L. johnsonii (Lj) (n=25 birds/pen, 8 pens/treatment).
All treatment groups with the exception of the Control group were challenged with C.
perfringens (Cp). Histological examination revealed that OA and Lj were unsuccessful in
preserving intestinal architecture, however ZnB maintained villus:crypt structure comparable
to the Control group. Total goblet cell (GC) number/mm villus surface length was not
significantly different amongst any of the dietary treatment groups. No interactions between
dietary treatment and mucin type were observed. The number of neutral mucin containing GC
was significantly greater in comparison to the number of total acidic mucin containing GC in
all dietary treatment groups. Cp challenge did not appear to affect the number of neutral or
acidic mucin GC however ZnB decreased the number of both neutral and total acidic mucin
GC. No significant differences in acidic mucin subtypes (sulphated, sialylated and
intermediate) were observed amongst dietary treatments. Sulphated mucin containing GC
were the dominant subtype within all treatment groups. Understanding how Cp affects the
intestinal wall will assist in the search for NE preventatives.
I. INTRODUCTION
Necrotic enteritis (NE), caused by Clostridium perfringens (Cp), is a serious disease
threatening the poultry industry worldwide (McDevitt et al., 2006). It is characterised by
mortality and intestinal lesions in the clinical form, whilst sub-clinical NE is associated with
diarrhoea and lowered feed efficiency (Van Immerseel et al., 2004). For many years NE has
primarily been controlled by the inclusion of in-feed sub-therapeutic antibiotics. The
withdrawal of in-feed antibiotics in the European Union and voluntary reductions in other
regions has increased the incidence of NE, hence there is a need to identify non-antibiotic
alternatives for NE prevention. The probiotic, L. johnsonii (Lj) and a commercial organic acid
(OA) blend have demonstrated potential as antibiotic alternatives (La Ragione et al; Zelenika
et al., 2005). Lj has been shown to suppress all aspects of colonisation of Cp, possibly by
immunomodulation or competitive exclusion (La Ragione et al., 2004), whilst OA has
improved bird performance, perhaps by disrupting the proton gradient across the bacterial
cell membrane (Zelenika et al., 2005). Probiotics and organic acids have also been shown to
alter the mucin profile during bacterial challenge or in unchallenged birds respectively.
The mucus layer is the first line of defence encountered by gut bacteria. Mucins,
which are the primary constituent of the mucus layer, are high molecular weight
1
School of Animal and Veterinary Sciences, The University of Adelaide, Roseworthy Campus, South Australia,
5371
2
SARDI, Pig and Poultry Production Institute, Roseworthy, South Australia, 5371
3
Faculty of Veterinary Medicine, University of Montreal, St-Hyacinthe, Québec, J2S 7C6, Canada
211
Aust. Poult. Sci. Symp. 2010...21
glycoproteins secreted by intestinal goblet cells (Smirnov et al., 2006). Goblet cells can be
histologically separated into two types; those that produce neutral or acidic mucin on the
basis of terminal sugars. The latter can be further divided into two subtypes; sulphated or
sialylated. Diet and bacteria are known to influence mucin subtype distribution, however little
is known in regard to changes in mucin profile during Cp challenge. It is well known that NE
disrupts intestinal integrity and villus:crypt architecture, consequently lowering nutrient
absorption and bird performance. In-feed antibiotics are known to reduce NE associated
intestinal damage however there is minimal literature on the ability of organic acids or
probiotics to prevent gut damage in NE challenged birds. This study investigated the effect of
Lj and an OA blend on the intestinal morphology and mucin profile of Cp challenged broilers
to aid evaluation of their potential as non-antibiotic preventatives for NE.
II. MATERIALS AND METHODS
This NE challenge experiment was carried out at the University of New England (Armidale,
NSW, Australia) under the guidance of Dr. Lene Mikkelsen and colleagues, as part of CRC
project 05-2 managed by SARDI researchers. A total of 1200 male Cobb 500 broilers were
reared from day old in 48 pens (n=25 birds/pen). Pens were randomly assigned to the
following six treatment groups (8 pens/group) unchallenged (Control), C. perfringens
challenged (Cp), ZnB (50 ppm in the diet), OA (2kg/ton in the diet), Lj (109 cfu/ml L.
johnsonii in PBS on days 1, 3, 7 and 12 administered by oral gavage) or vehicle (PBS gavage
on corresponding days). All treatment groups with the exception of the Control were
inoculated with a suspension of three Eimeria sp. on day 9 and Cp type A (3.5 x 108 cfu/ml)
on day 15, according to the NE challenge procedure described previously (Kocher et al.,
2004; Mikkelsen et al., 2009). Birds were fed a starter diet and corresponding dietary
treatments from days 0-6 and 15-22 inclusive and a finisher diet until day 28. Between days
7-15 birds were fed a high protein diet (50% fishmeal and 50% starter) to aid NE
development. On day 18, l cm tissue samples were collected from the midpoint of the
duodenum, jejunum and ileum (n=12 birds/treatment) and later embedded in paraffin.
Standard haematoxylin and eosin staining was performed on 8 μm jejunal sections.
Villus height (VH), crypt depth (CD) and villus width (VW) were measured from 15
randomly selected villi and associated crypts from each bird by microscopic examination.
Apparent villus surface area (SA) was calculated using the formula SA=2π(VW/2)(VH)
(Sakamoto et al., 2000).
Periodic acid-Schiff (PAS) staining methods were used to identify neutral mucin
containing GC and high iron diamine-alcian blue (HID-AB) pH 2.5 staining was used to
distinguish between the two acidic mucin subtypes; sialylated and sulphated, on 10 randomly
selected villi for each bird. GC containing both sulphated and sialylated mucin concurrently
were termed intermediate. Light microscopy and image analysis were used to determine the
mean number of GC per unit of villus surface length (mm).
III. RESULTS
Birds from the Control and ZnB treatment groups had significantly greater villus heights and
VH:CD and reduced crypts and villus width (P < 0.05; Table 1) in comparison to birds from
the Cp, OA, Vehicle and Lj treatment groups. ZnB fed birds maintained an intestinal
villus:crypt architecture comparable to Control birds (P > 0.05; Table 1). No significant
differences in any of the intestinal histomorphological parameters were observed amongst the
Cp, OA, Vehicle and Lj treatment groups with the exception of the OA group displaying a
significantly greater mean villus width in comparison to the Cp group (P < 0.05; Table 1).
212
Aust. Poult. Sci. Symp. 2010...21
Table 1
Histomorphological parameters in the jejunum from 18 day old broilers, 3
days post Clostridium perfringens challenge1
Villus height
(μm)
Crypt depth
(μm)
VH:CD
Villus width
(μm)
Control
880 ± 39.4 x
182 ± 12.3 y
5.3 ± 0.4 x
68 ± 4.5 z
Apparent
villus SA2
(x105 μm2)
1.9 ± 0.1
C. perfringens
656 ± 64.2 y
306 ± 14.7 x
2.3 ± 0.3 y
85 ± 2.5 y
1.8 ± 0.2
Zinc bacitracin
997 ± 42.4 x
221 ± 10.2 y
4.9 ± 0.3 x
73 ± 1.9 z
2.3 ± 0.1
Organic acid
550 ± 66.7 y
298 ± 17.0 x
2.1 ± 0.2 y
97 ± 2.9 x
1.7 ± 0.2
Vehicle
576 ± 61.7 y
308 ± 18.5 x
2.1 ± 0.3 y
92 ± 2.8 xy
1.7 ± 0.2
L. johnsonii
641 ± 60.4 y
307 ± 12.4 x
2.3 ± 0.2 y
93 ± 2.7 xy
1.8 ± 0.2
Values are means ± SEM (n=12 birds/treatment). Means within the same column with different
superscripts differ significantly (P < 0.05) 2 SA; surface area.
1
No significant difference in total GC number was observed amongst treatment groups (P >
0.05; Table 2). No interaction amongst dietary treatments and mucin type was evident. Cp
challenge did not appear to affect the number of neutral or total acidic mucin GC; whilst ZnB
decreased the number of neutral mucin GC in comparison to birds in the Cp group and the
number of total acidic mucin GC in comparison to birds in the Control, Cp and Lj group. The
number of neutral mucin containing GC was significantly greater than total acidic mucin
containing GC in all dietary treatment groups (P < 0.05; Table 2).
Table 2
Goblet cell (GC) mucin composition and total GC numbers in the jejunum of
18 day old broilers, 3 days post Clostridium perfringens challenge1
Neutral
Total acidic
Total GC
Control
75.0 ± 3.1xy
65.0 ± 2.0x
139.8 ± 4.4
C. perfringens
80.6 ± 3.1x
63.6 ± 4.8xy
144.2 ± 7.2
Zinc bacitracin
68.8 ± 2.2y
56.8 ± 1.8z
125.6 ± 3.7
Organic acid
74.4 ± 2.1xy
59.3 ± 1.6yz
133.7 ± 2.5
Vehicle
73.1 ± 2.7y
58.7 ± 3.1yz
131.9 ± 5.2
L. johnsonii
75.0 ± 2.3
62.5 ± 2.6
137.5 ± 4.0
xy
xy
Values are means ± SEM of the number of goblet cells/mm of villus surface length (n=12
birds/treatment). Means within the same column with different superscripts differ significantly (P <
0.05).
1
There were no significant differences in acidic mucin subtype amongst dietary treatments (P
> 0.05; Table 3). However the number of acidic mucin subtypes was significantly different
within each dietary treatment (P < 0.05; Table 3). Sulphated mucin was the dominant mucin
subtype within all dietary treatments; whilst numbers of intermediate mucin containing GC
were the lowest (P < 0.05; Table 3).
213
Aust. Poult. Sci. Symp. 2010...21
Table 3
Acidic goblet cell mucin composition in the jejunum of 18 day old broilers, 3
days post Clostridium perfringens challenge1
Sulphated
Sialylated
22.3 ±
4.4
b
Intermediate
11.0 ±
Control
32.6 ±
5.2
a
C. perfringens
40.0 ±
6.2
a
16.1 ±
3.3
b
7.5 ±
1.6
c
Zinc bacitracin
35.4 ±
4.6
a
12.8 ±
3.4
b
8.6 ±
1.5
c
Organic acid
33.0 ±
5.0
a
18.9 ±
3.6
b
7.4 ±
1.3
c
Vehicle
29.0 ±
4.4
a
20.4 ±
4.0
b
9.3 ±
1.3
c
L. johnsonii
33.0 ±
5.4
a
22.9 ±
4.9
b
6.7 ±
1.1
c
1.3
c
Values are means ± SEM of the number of goblet cells/mm of villus surface length (n=12
birds/treatment). Means within the same row with different superscripts differ significantly (P < 0.05).
1
IV. CONCLUSION
The probiotic, L. johnsonii and an organic acid blend did not successfully ameliorate
histological signs of intestinal damage or alter the intestinal mucin profile in broilers
challenged with Cp. Conventional treatment with zinc bacitracin maintained intestinal
villus:crypt architecture in Cp challenged birds. The mucin profile did not appear to be
affected by Cp challenge, however zinc bacitracin decreased the number of neutral and total
acidic mucin containing GC. Neutral mucin was the dominant mucin type contained in GC
irrespective of treatment, whilst sulphated mucin was the primary acidic mucin GC subtype.
Thus it is possible that Cp does not influence the mucin profile however further investigation
is required, particularly into the function and structure of acidic mucin subtypes. Although L.
johnsonii and the organic acid blend have not demonstrated potential as antibiotic alternatives
in this study, similar compounds and organisms, which can prevent Cp colonisation and
maintain gut integrity, may improve bird performance and prevent NE.
REFERENCES
Kocher A, Choct M, Teo A, Tan HM, Carter RR (2004) Proceedings, Australian Poultry
Science Symposium 16, 84-87.
La Ragione RM, Narbad A, Gasson MJ, Woodward MJ (2004) Letters of Applied
Microbiology 38,197-205.
McDevitt RM, Brooker JD, Acamovic T, Sparks NHC (2006) World’s Poultry Science
Journal 62, 221-247.
Mikkelsen LL, Vidanarachchi JK, Olnood CG, Bao YM, Selle PH, Choct M (2009) British
Poultry Science 50, 66-75.
Sakamoto K, Hirose H, Onizuka A, Hayashi M, Futamura N, Kawamura Y, Ezaki T (2000)
Journal of Surgical Research 94, 99-106.
Smirnov A, Tako E, Ferket PR, Uni Z (2006) Poultry Science 85, 669-673.
Van Immerseel F, De Buck J, Pasmans F, Huyghebaert G, Haesebrouck F, Ducatelle R
(2004) Avian Pathology 33, 537-549.
Zelenika TA, Mikec M, Lisjak M, Novak IL, Kos K, Valentic A (2005) Croatian Veterinary
Institute, Poultry Centre.
214
AUTHOR INDEX
Name
Page(s)
Allison, G.E.
Ayalew, W.
Balding, K.
Barnett, J.L.
Bhuiyan, M.M.
Black, I.D.
Black, J.L.
Blok, M.
Borg, S.S.
Borland, E.A.
Boulianne, M.
Brown, P.R
Bryden, W.L.
Burgess, S.K.
Cadogan, D.J.
Carroll, S.
Chauynarong, N
Cheetham, B.F.
Choct, M.
Chousalkar, K.K.
Cronin, G.M.
De Beer, M.
de Lange, L.L
de Reu, K.
Dubois, S
Downing, J.A.
Dunlop, M.
Easey, P.
Fike, S.
Flinn, P.C.
Forder, R.E.A.
Frio, A.
Geier, M.S.
Gibson, R.A
Gill, R.J
Glatz, P.C.
Golder, H.M.
Gous, R.M
Gruspeerdt, K.
Hangoor, E.
Hazel, S.
207
145
195
130
99
145
51
44
130
134
207, 211
178
94
118
50, 94,
95
136
107
50, 207
86, 111
85, 130, 190
103
1, 191
74, 126
26
68, 130, 182, 186, 190
122
118
174
51
211
174
51, 64, 207, 211
64
68
134, 135, 145,
211
36, 199
74, 126
90
134
Email Address
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
Hazeldene, J.
Hemsworth, P.H.
Herman, L.
Hetland, H.
Heyndrickx, M.
Huang, H.K.
Hughes, R.J.
Hynd, P.I.
Iji, P.A.
Ischan, J.
Islam, A.F.M.F.
Jansen, T.
Jansman, A.
Jarman, T.
Kanto, U.
Kartikasari, L.R.
Kemp, C.
Kocher, A.
Kwakernaak, C.
Laine, S.M.
Leary, A.
Lemme, A.
Li, L.
Li. X.
MacAlpine, R.
Mack, S.D.
Madsen, T.G.
Makrides, M
Manu, V.
Martin, C
Messens, W.
Miao, Z.H.
Mikkelsen, L.L.
Molan, A.L.
Naylor, A
Nichol, R.
Nielsen, S.G.
Noblet, J.
Nolan, J.V.
Olnood, C.
Ophel-Keller, K.
Pandi, J.K.
Parker, R.
Perez-Maldonado, R.
60
85
74, 126
82
74, 126
103
51, 64, 145, 195, 207, 211
211
72, 99, 136
186
99, 118, 122
145
44
174
136
64
95
60, 141, 174
44
85
141, 174
90, 95
115
94
195
161
90, 95
64
145
60
74, 126
134, 135, 145
99, 195, 207
203
60
199
51
26
72
207
195
145
171
55, 140, 195
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
Petherick, J.C.
Pym, R.
Ravindran, G.
Ravindran, V.
Renz, K.G
Rimmington, H.
Roberts, J.R.
Rodda, B.K.
Rodenburg, B.
Rodrigues, H.
Selle, P.H.
Setia, A.
Sparrla, J.K.W.M.
Steiner, T.
Storey, T.H.
Sultan. A.
Svihus, B.
Tancharoenrat, P
Taylor, W.
Thomas, D.V.
Timmons, B
Torok, V.A.
Tredrea, A.M
Treloar, T.
Tukei, A.
Uyttendaele, M.
van der Klis, J D
van Milgen, J.
Wahanui, J.
Walkden-Brown, S.W.
Wells, B.
Whitehead, C.C
Widodo, A.
Wijtten, P.J.A.
Williams, K.J.
Wilson, T.
Wyatt, S.C.
Zaefarian, F.
Zhang, D.
85
153
59
59, 141, 203
107
134, 135,
86, 111
134, 135, 145
74
55
68
207
90
199
130
94
9, 82
59
182
141, 203
174
195, 207
51
140
50
74, 126
17, 44
26
145
107, 118, 122
118, 122
22
72
90, 191
190
60
134, 135
59
94
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]